import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import mpl_toolkits as mplot3d
import scipy as sp
pd.set_option("display.max_columns", None)
from tqdm import tqdm
tqdm.pandas()

def lighten_color(color, amount=0.5):
    """
    Lightens the given color by multiplying (1-luminosity) by the given amount.
    Input can be matplotlib color string, hex string, or RGB tuple.

    Examples:
    >> lighten_color('g', 0.3)
    >> lighten_color('#F034A3', 0.6)
    >> lighten_color((.3,.55,.1), 0.5)
    """
    import matplotlib.colors as mc
    import colorsys
    try:
        c = mc.cnames[color]
    except:
        c = color
    c = colorsys.rgb_to_hls(*mc.to_rgb(c))
    return colorsys.hls_to_rgb(c[0], 1 - amount * (1 - c[1]), c[2])

d0 = pd.read_excel("risk literacy_985 responses.xlsx", sheet_name = "643 above median").rename({"#":"ResponseId"}, axis = 1)
d0 = d0.loc[ (d0["Please state your current occupation."] != "Investment Professional, i.e. stock broker/trader; financial planner/advisor; portfolio manager; investment banker; stock analyst; venture capital/private equity; insurance agent, etc.") ]
d0

d0["Please state your current occupation."].value_counts()

Please state your current occupation.
Employee/Consultant in Other than the Finance Sector    378
Entrepreneur or Own Business                             96
Student                                                   7
Home Maker or not employed                                5
Retired person                                            3
Name: count, dtype: int64

# Age groups/rec = [2,3,4]
# Age group 2 = 18 to 35 y/o
# Age group 3 = 36 to 55 y/o
# Age group 4 = 56 to 75 y/o (75 y/o, i.e, within the scope of the data we have, it can mean 60 and above also)

d0["age_rec"] = np.where( d0["Please mark your age (in years)"] == "18 - 25" , 2,
                          np.where(d0["Please mark your age (in years)"] == "25 - 35", 2,
                                   np.where( d0["Please mark your age (in years)"] == "36 - 45", 3,
                                            np.where( d0["Please mark your age (in years)"] == "46 - 55", 3,
                                                     np.where( d0["Please mark your age (in years)"] == "56 - 65", 4,
                                                              np.where( d0["Please mark your age (in years)"] == "Above 65", 4, 0
                                                                      )
                                                             )
                                                    )
                                           )
                                  )
                        )

d0["age"] = np.where( d0["Please mark your age (in years)"] == "18 - 25" , (18 + 25) / 2,
                          np.where(d0["Please mark your age (in years)"] == "25 - 35", (26 + 35) / 2,
                                   np.where( d0["Please mark your age (in years)"] == "36 - 45", (36 + 45) / 2,
                                            np.where( d0["Please mark your age (in years)"] == "46 - 55", (46 + 55) / 2,
                                                     np.where( d0["Please mark your age (in years)"] == "56 - 65", (56 + 65) / 2,
                                                              np.where( d0["Please mark your age (in years)"] == "Above 65", (66 + 75) / 2, 0
                                                                      )
                                                             )
                                                    )
                                           )
                                  )
                        )


# Secondary upto 10 – ISCED 3
# Senior Secondary upto 12 - ISCED 3 
# Diploma and voca – ISCED 2
# UG – ISCED 2
# PG – ISCED 1
# PhD and higher – ISCED 1

d0["isced"] = np.where(d0["What is currently your highest Education?"] == "Post Graduate", 1,
                        np.where(d0["What is currently your highest Education?"] == "Under Graduate", 2,
                                 np.where(d0["What is currently your highest Education?"] == "Ph.D. or higher", 1,
                                          np.where(d0["What is currently your highest Education?"] == "Diploma or vocation training", 2,
                                                   np.where(d0["What is currently your highest Education?"] == "School degree (X or XII)", 3, 0
                                                           )
                                                  )
                                         )
                                )
                       )

d0["income"] = np.where(d0["Please indicate your approximate annual personal income from all sources for last year"] == "&lt; 5,00,000", 1,
                         np.where(d0["Please indicate your approximate annual personal income from all sources for last year"] == "5,00,000 - 15,00,000", 2,
                                  np.where(d0["Please indicate your approximate annual personal income from all sources for last year"] == "15,00,001 - 25,00,000", 3,
                                           np.where(d0["Please indicate your approximate annual personal income from all sources for last year"] == "25,00,001 - 35,00,000", 4,
                                                    np.where(d0["Please indicate your approximate annual personal income from all sources for last year"] == "35,00,001 - 45,00,000", 4,
                                                             np.where( d0["Please indicate your approximate annual personal income from all sources for last year"] == "&gt; 45,00,000", 5,0
                                                                     )
                                                            )
                                                   )
                                           )
                                  )
                         )

d0["wealth"] = d0["Please provide a rough guess (in Indian Rupees) of the worth of your household's assets. Please do not forget to correct it for your debts, such as a mortgage or any loans you might have."]

matchCol = ["q8_2_1", "q8_2_2", "q8_2_3", "q8_2_4", "q8_2_5", "q8_3", "q8_4", "q8_5berlin_1", 
            "q8_5london_1", "q8_5paris_1", "q8_6", "q8_7", "q9_1_1", "q9_2_1", "q9_3",
            "q10_1_1", "q10_2_1", "q10_3_1", "q10_4"]
d0[ matchCol ] = 0

d0.columns

Index(['ResponseId', 'Please indicate your Gender.',
       'Please mark your age (in years)',
       'What is currently your highest Education?',
       'Please state your current occupation.',
       'How do you describe your willingness to take financial risk in general?',
       'Given the number of years that you have held various investments and the amount of investing you might have done, what degree of investment experience in the stock market do you have?',
       'If an expert tries to worry or scare me, i.e. a financial advisor about my financial situation, I choose another expert.',
       'I only buy a financial product I understand.', 'I trust doctors.',
       'When I want to buy a bigger item like a refrigerator or an expensive item of clothing, I wait a month to see whether I still want it and only buy it then.',
       'I always keep in mind that everything I do on the web could be used to my disadvantage.',
       'In my household, we/I spend:', 'In my household, we/I',
       '_Distribution 1_\n\nHow risky do you perceive the investment to be?',
       '_Distribution 2_\n\nHow risky do you perceive the investment to be?',
       '_Distribution 3_\n\nHow risky do you perceive the investment to be?',
       '_Distribution 4_\n\nHow risky do you perceive the investment to be?',
       '_Distribution 5_\n\nHow risky do you perceive the investment to be?',
       'Distribution 6\n\nHow risky do you perceive the investment to be?',
       'Distribution 7\n\nHow risky do you perceive the investment to be?',
       'Distribution 8\n\nHow risky do you perceive the investment to be?',
       '*Mumbai * A = 9 out of 10000   *OR   * B = 1 out of 1000',
       '*Bengaluru * A = 0.7%   *OR  * B = 0.099%',
       '*Kolkata * A = 0.61%   *OR   * B = 6 out of 10000', 'HIV test',
       'Fingerprint', 'DNA test', 'Cancer screening test',
       'Professional horoscope',
       'A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%.  Which of the following statements is true?',
       'There is an official prediction that the national stock market will grow 2% annually over the next 5 years.  This means that…',
       'Imagine you are told that the price of the stock Soya Ruchi increases from INR 60 to INR 120 after the company merger. What does this mean?',
       'It is predicted that Indigo Bank has 30% chance of default next year. Which of the following alternatives is the most appropriate interpretation of the statement?',
       'The probability that the economy will go into a recession this year is 30%. If the economy goes into recession, the probability that the stock market will decrease is 80%. If the economy does not go into a recession, the probability that the stock market will decrease is 23%. What is the probability that the economy goes into recession given that the stock market decreased?',
       'A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:',
       'Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000.  This implies:',
       'Imagine that we flip a fair coin 1,000 times. What is your best guess about how many times the coin will come up heads in 1,000 flips? \n\n\_\_\_\_\_\_ times out of 1,000.',
       'In the Bingo Lottery, the chance of winning a $10 prize is 1%. What is your best guess about how many people will win a $10 prize if 1,000 people each buy a single ticket for Bingo Lottery?\n\n\_\_\_\_\_\_ person(s) out of 1,000.',
       'In a sweepstakes, the chance of winning a car is 1 in 1,000. What percentage of tickets for the sweepstakes wins a car?\n\n\_\_\_\_\_ % of tickets',
       'About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]',
       'Approximately what percentage (%) of people who die from cancer die from colon cancer, breast cancer, and prostate cancer taken together?',
       'The following figure shows the number of men and women among a group of smartphone users. The total number of circles is 100. \n\nHow many more men than women are there among the 100 people using a smartphone?',
       'In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?',
       'Please indicate your approximate annual personal income from all sources for last year',
       'Please provide a rough guess (in Indian Rupees) of the worth of your household's assets. Please do not forget to correct it for your debts, such as a mortgage or any loans you might have.',
       'uid', 'Response Type', 'Start Date (UTC)', 'Stage Date (UTC)',
       'Submit Date (UTC)', 'Network ID', 'LTA', 'age_rec', 'age', 'isced',
       'income', 'wealth', 'q8_2_1', 'q8_2_2', 'q8_2_3', 'q8_2_4', 'q8_2_5',
       'q8_3', 'q8_4', 'q8_5berlin_1', 'q8_5london_1', 'q8_5paris_1', 'q8_6',
       'q8_7', 'q9_1_1', 'q9_2_1', 'q9_3', 'q10_1_1', 'q10_2_1', 'q10_3_1',
       'q10_4'],
      dtype='object')

d0["HIV test"].value_counts()

HIV test
Yes    340
No     149
Name: count, dtype: int64

d0["q8_2_1"] = np.where(d0["HIV test"] == "Yes", 1, 2)
d0["q8_2_2"] = np.where(d0["Fingerprint"] == "Yes", 1, 2)
d0["q8_2_3"] = np.where(d0["DNA test"] == "Yes", 1, 2)
d0["q8_2_4"] = np.where(d0["Cancer screening test"] == "Yes", 1, 2)
d0["q8_2_5"] = np.where(d0["Professional horoscope"] == "Yes", 1, 2)

d0["q8_3"] = np.where(d0["A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%.  Which of the following statements is true?"] == "The lower the quality of the study, the more likely that future studies will change the risk estimate.", 1,
                          np.where(d0["A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%.  Which of the following statements is true?"] == "The higher the quality of the study, the more likely that future studies will change the risk estimate.", 2,
                                  np.where(d0["A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%.  Which of the following statements is true?"] == "Irrespective of the quality of the study, future studies will not change the risk estimate.", 3,
                                          np.where(d0["A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%.  Which of the following statements is true?"] == "Irrespective of the quality of the study, future studies will change the risk estimate substantially anyway.", 4,0
                                                  )
                                          )
                                  )
                         )


d0["q8_4"] = np.where(d0["There is an official prediction that the national stock market will grow 2% annually over the next 5 years.  This means that…"] == "the growth rate will be 0.4% on average each year", 1,
                          np.where(d0["There is an official prediction that the national stock market will grow 2% annually over the next 5 years.  This means that…"] == "the growth rate over five years will be exactly 2%", 2,
                                  np.where(d0["There is an official prediction that the national stock market will grow 2% annually over the next 5 years.  This means that…"] == "the growth rate over five years will be between 1% and 3%", 3,
                                          np.where(d0["There is an official prediction that the national stock market will grow 2% annually over the next 5 years.  This means that…"] == "it is not possible to predict the growth rate with certainty.", 4,0
                                                  )
                                          )
                                  )
                         )

d0["q8_5berlin_1"] = np.where(d0["*Mumbai * A = 9 out of 10000   *OR   * B = 1 out of 1000"] == "A", 1, 2)
d0["q8_5london_1"] = np.where(d0["*Bengaluru * A = 0.7%   *OR  * B = 0.099%"] == "A", 1, 2)
d0["q8_5paris_1"] = np.where(d0["*Kolkata * A = 0.61%   *OR   * B = 6 out of 10000"] == "A", 1, 2)

d0["q8_6"] = np.where(d0["Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000.  This implies:"] == "The medication increases recovery by 100%", 1,
                          np.where(d0["Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000.  This implies:"] == "The medication increases recovery by 50%", 2,
                                   np.where(d0["Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000.  This implies:"] == "The medication increases recovery by 2%", 3,
                                            np.where(d0["Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000.  This implies:"] == "None of the above is implied.", 4, 0
                                                    )
                                           )
                                  )
                         )

d0["q8_7"] = np.where(d0["A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:"] == "5 in 100 people prior to the intervention to 6 out of 100 people after the intervention", 1,
                          np.where(d0["A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:"] == "100 in 10,000 people prior to the intervention to 120 out of 10,000 people after the intervention", 2,
                                   np.where(d0["A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:"] == "70 in 100 people prior to the intervention to 90 out of 100 people after the intervention", 3,
                                            np.where(d0["A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:"] == "it is not possible to determine which of the answers is correct given the information provided", 4, 0
                                                    )
                                           )
                                  )
                         )


d0["q9_1_1"] = d0["Approximately what percentage (%) of people who die from cancer die from colon cancer, breast cancer, and prostate cancer taken together?"].copy()
d0["q9_2_1"] = d0["The following figure shows the number of men and women among a group of smartphone users. The total number of circles is 100. \n\nHow many more men than women are there among the 100 people using a smartphone?"].copy()

d0["q9_3"] = np.where(d0["In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?"] == "Crosicol", 1,
                          np.where(d0["In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?"] == "Hertinol", 2,
                                   np.where(d0["In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?"] == "They are equal", 3,
                                            np.where(d0["In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?"] == "Can’t say", 4, 0
                                                    )
                                           )
                                  )
                         )

d0["q10_1_1"] = d0["Imagine that we flip a fair coin 1,000 times. What is your best guess about how many times the coin will come up heads in 1,000 flips? \n\n\_\_\_\_\_\_ times out of 1,000."].copy()
d0["q10_2_1"] = d0["In the Bingo Lottery, the chance of winning a $10 prize is 1%. What is your best guess about how many people will win a $10 prize if 1,000 people each buy a single ticket for Bingo Lottery?\n\n\_\_\_\_\_\_ person(s) out of 1,000."].copy()
d0["q10_3_1"] = d0["In a sweepstakes, the chance of winning a car is 1 in 1,000. What percentage of tickets for the sweepstakes wins a car?\n\n\_\_\_\_\_ % of tickets"].copy()

d0["q10_4"] = np.where(d0["About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]"] == "9 out of 59", 1,
                          np.where(d0["About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]"] == "9 out of 10", 2,
                                   np.where(d0["About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]"] == "59 out of 1000", 3,
                                            np.where(d0["About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]"] == "59 out of 100", 4, 0
                                                    )
                                           )
                                  )
                         )

scoreColumns = ["certainty1", "certainty2", "certainty3", "certainty4", "certainty5", "uncertainty1", "uncertainty2", "numeracy1", "numeracy2", "numeracy3", "numeracy4", "numeracy5", "graph1", "graph2", "graph3", "riskcomprehension1", "riskcomprehension2", "riskcomprehension3", "riskcomprehension4", "riskcomprehension5", "bayesian1"]
d0[scoreColumns] = 0

# Assigning scores

def scoring1(surveyFacet):
    surveyFacet.loc[ surveyFacet["q8_2_1"] == 2, "certainty1"] = 1
    surveyFacet.loc[ surveyFacet["q8_2_2"] == 2, "certainty2"] = 1
    surveyFacet.loc[ surveyFacet["q8_2_3"] == 2, "certainty3"] = 1
    surveyFacet.loc[ surveyFacet["q8_2_4"] == 2, "certainty4"] = 1
    surveyFacet.loc[ surveyFacet["q8_2_5"] == 2, "certainty5"] = 1

    surveyFacet.loc[ surveyFacet["q8_3"] == 1, "uncertainty1"] = 1
    surveyFacet.loc[ surveyFacet["q8_4"] == 4, "uncertainty2"] = 1

    surveyFacet.loc[ surveyFacet["q8_5berlin_1"] == 2, "riskcomprehension1"] = 1
    surveyFacet.loc[ surveyFacet["q8_5london_1"] == 1, "riskcomprehension2"] = 1
    surveyFacet.loc[ surveyFacet["q8_5paris_1"] == 1, "riskcomprehension3"] = 1
    
    surveyFacet.loc[ surveyFacet["q8_6"] == 1, "riskcomprehension4"] = 1
    surveyFacet.loc[ surveyFacet["q8_7"] == 4, "riskcomprehension5"] = 1

    surveyFacet.loc[ surveyFacet["q9_1_1"] == 25, "graph1"] = 1
    surveyFacet.loc[ surveyFacet["q9_2_1"] == 20, "graph2"] = 1
    surveyFacet.loc[ surveyFacet["q9_3"] == 3, "graph3"] = 1

    surveyFacet.loc[ surveyFacet["q10_1_1"] == 500, "numeracy1"] = 1
    surveyFacet.loc[ surveyFacet["q10_2_1"] == 10, "numeracy2"] = 1
    surveyFacet.loc[ (surveyFacet["q10_3_1"] == 0.1) | (surveyFacet["q10_3_1"] == ".1") | (surveyFacet["q10_3_1"] == ",1") , "numeracy3"] = 1
    surveyFacet.loc[ surveyFacet["q10_4"] == 1, "bayesian1"] = 1

    # Assigning total scores
    # surveyFacet["Certainty score_5"] = surveyFacet["certainty1"] + surveyFacet["certainty2"] + surveyFacet["certainty3"] + surveyFacet["certainty4"] + surveyFacet["certainty5"]
    #surveyFacet["Uncertainty score_5"] = surveyFacet["uncertainty1"] + surveyFacet["uncertainty2"]

    #surveyFacet["Number Comprehension score_5"] = surveyFacet["numeracy1"] + surveyFacet["numeracy2"] + surveyFacet["numeracy3"] + surveyFacet["numeracy4"] + surveyFacet["numeracy5"]
    #surveyFacet["Graph Comprehension score_5"] = surveyFacet["graph1"] + surveyFacet["graph2"] + surveyFacet["graph3"]

    #surveyFacet["Calculation score_4"] = surveyFacet["riskcalculation1"] + surveyFacet["riskcalculation2"] + surveyFacet["riskcalculation3"] + surveyFacet["riskcalculation4"]

    #surveyFacet["Total Score_19"] = surveyFacet["Certainty score_5"] + surveyFacet["Uncertainty score_5"] + surveyFacet["Number Comprehension score_5"] + surveyFacet["Graph Comprehension score_5"] + surveyFacet["Calculation score_4"]

    return surveyFacet

d0 = d0.groupby(["ResponseId"]).progress_apply(scoring1)
d0

100%|███████████████████████████████████████████████████████████████████████████████| 489/489 [00:02<00:00, 167.29it/s]

def scoring2(survey):
    
    survey["Certainty_5"] = survey["certainty1"] + survey["certainty2"] + survey["certainty3"] + survey["certainty4"] + survey["certainty5"]
    
    survey["Uncertainty_2"] = survey["uncertainty1"] + survey["uncertainty2"]
    
    survey["RiskComprehension_5"] = survey["riskcomprehension1"] + survey["riskcomprehension2"] + survey["riskcomprehension3"] + survey["riskcomprehension4"] + survey["riskcomprehension5"]
    
    survey["GraphLiteracy_3"] = survey["graph1"] + survey["graph2"] + survey["graph3"]
    
    survey["Numeracy_2"] = survey["numeracy1"] + survey["numeracy2"] # + survey["numeracy3"]
    
    survey["Bayesianreasoning_1"] = survey["bayesian1"]
    
    survey["TotalScore_18"] = survey["Certainty_5"] + survey["Uncertainty_2"] + survey["RiskComprehension_5"] + survey["GraphLiteracy_3"] + survey["Numeracy_2"] + survey["Bayesianreasoning_1"]
    
    survey["Certainty_%"] = survey["Certainty_5"] / 5 * 100
    survey["Uncertainty_%"] = survey["Uncertainty_2"] / 2 * 100
    survey["RiskComprehension_%"] = survey["RiskComprehension_5"] / 5 * 100
    survey["GraphLiteracy_%"] = survey["GraphLiteracy_3"] / 3 * 100
    survey["Numeracy_%"] = survey["Numeracy_2"] / 2 * 100
    survey["Bayesianreasoning_%"] = survey["Bayesianreasoning_1"] / 1 * 100
    
    survey["TotalScore_%"] = survey["TotalScore_18"] / 18 * 100
    
    
    colReq = ["ResponseId", "age", "age_rec", "isced", "income", "wealth", "Certainty_5", "Uncertainty_2", "RiskComprehension_5", "GraphLiteracy_3",
              "Numeracy_2", "Bayesianreasoning_1", "Certainty_%", "Uncertainty_%", "RiskComprehension_%", "GraphLiteracy_%", "Numeracy_%", "Bayesianreasoning_%",
              "TotalScore_18", "TotalScore_%",]
    
    survey1 = survey[colReq].copy()
    return survey1

d01 = scoring2(d0)
d01

d1 = pd.read_csv('Indian Risk Survey_Besample_Filtered_12_2024.csv')
d11 = d1.iloc[:, 19:62].copy()
d11.insert(0, "ResponseId",0)
d11["ResponseId"] = d1["ResponseId"].copy()
d11 = d11.loc[ (d11["Q4"] != "Stock analyst") & (d11["Q4"] != "Insurance agent, etc") & (d11["Q4"] != "Venture capital/private equity")]

d11["Q4"].value_counts()

Q4
Salaried/Employee/Consultant in a sector other than Finance    167
Student                                                        100
Not employed                                                    50
Entrepreneur/Business Owner in a sector other than Finance      30
Retired                                                          3
Name: count, dtype: int64

d11["age"] = d11["Q2"]

# Age groups/rec = [2,3,4]
# Age group 2 = 18 to 35 y/o
# Age group 3 = 36 to 55 y/o
# Age group 4 = 56 to 75 y/o (75 y/o, i.e, within the scope of the data we have, it can mean 60 and above also)

d11["age_rec"] = np.where( (d11["age"] >= 18) & (d11["age"] <= 35), 2,
                          np.where( (d11["age"] >= 36) & (d11["age"] <= 55), 3,
                                   np.where( (d11["age"] >= 56), 4, 0
                                           )
                                  )
                         )


# Secondary upto 10 – ISCED 3
# Senior Secondary upto 12 - ISCED 3 
# Diploma and voca – ISCED 2
# UG – ISCED 2
# PG – ISCED 1
# PhD and higher – ISCED 1

d11["isced"] = np.where(d11["Q3"] == "Post-Graduate Program", 1,
                        np.where(d11["Q3"] == "Undergraduate Program", 2,
                                 np.where(d11["Q3"] == "Ph.D. and higher", 1,
                                          np.where(d11["Q3"] == "Diploma and Vocational Training", 2,
                                                   np.where(d11["Q3"] == "Secondary School (11th to 12th Std.)", 3,
                                                            np.where( d11["Q3"] == "Primary School (up to 10th Std.)", 3,
                                                                     np.where( d11["Q3"] == "M.Phil.", 1, 0
                                                                             )
                                                                    )
                                                           )
                                                  )
                                         )
                                )
                       )

d11["income"] = np.where(d11["Q15a"] == "< INR 500,000", 1,
                         np.where(d11["Q15a"] == "INR 500,001 – INR 15,00,000", 2,
                                  np.where(d11["Q15a"] == "INR 1500,001 – INR 30,00,000", 3,
                                           np.where(d11["Q15a"] == "INR 30,00,001 – INR 50,00,000", 4,
                                                    np.where(d11["Q15a"] == "INR 50,00,001 – INR 75,00,000", 5,
                                                             np.where( d11["Q15a"] == "> INR 75,00,000", 5,0
                                                                     )
                                                            )
                                                   )
                                           )
                                  )
                         )

d11["wealth"] = d11["Q14b.1"]

matchCol = ["q8_2_1", "q8_2_2", "q8_2_3", "q8_2_4", "q8_2_5", "q8_3", "q8_4", "q8_5berlin_1", 
            "q8_5london_1", "q8_5paris_1", "q8_6", "q8_7", "q9_1_1", "q9_2_1", "q9_3",
            "q10_1_1", "q10_2_1", "q10_3_1", "q10_4"]
d11[ matchCol ] = 0

d11.columns

Index(['ResponseId', 'Q0', 'Q1', 'Q2', 'Q3', 'Q4', 'Q5_1', 'Q6_1', 'Q8_1',
       'Q8_2', 'Q8_3', 'Q8_4', 'Q8_5', 'Q8_6', 'Q8_7', 'Q8_8', 'Q9b_1',
       'Q9b_4', 'Q9b_5', 'Q9b_6', 'Q9b_7', 'Q9b_8', 'Q9b_9', 'Q11a_1',
       'Q11a_2', 'Q11a_3', 'Q11b', 'Q11c', 'Q11d', 'Q11h', 'Attention Check',
       'Q11i', 'Q12a', 'Q12b', 'Q12c', 'Q13a', 'Q13b', 'Q13c', 'Q13d', 'Q14a',
       'Q14b', 'Q14c', 'Q15a', 'Q14b.1', 'age', 'age_rec', 'isced', 'income',
       'wealth', 'q8_2_1', 'q8_2_2', 'q8_2_3', 'q8_2_4', 'q8_2_5', 'q8_3',
       'q8_4', 'q8_5berlin_1', 'q8_5london_1', 'q8_5paris_1', 'q8_6', 'q8_7',
       'q9_1_1', 'q9_2_1', 'q9_3', 'q10_1_1', 'q10_2_1', 'q10_3_1', 'q10_4'],
      dtype='object')

d11["q8_2_1"] = np.where(d11["Q11b"].str.contains("HIV test"), 1, 2)
d11["q8_2_2"] = np.where(d11["Q11b"].str.contains("Fingerprint"), 1, 2)
d11["q8_2_3"] = np.where(d11["Q11b"].str.contains("DNA test"), 1, 2)
d11["q8_2_4"] = np.where(d11["Q11b"].str.contains("Cancer screening test"), 1, 2)
d11["q8_2_5"] = np.where(d11["Q11b"].str.contains("Professional horoscope"), 1, 2)

d11["q8_3"] = np.where(d11["Q11c"] == "The lower the quality of the study, the more likely that future studies will change the risk estimate.", 1,
                          np.where(d11["Q11c"] == "The higher the quality of the study, the more likely that future studies will change the risk estimate.", 2,
                                  np.where(d11["Q11c"] == "Irrespective of the quality of the study, future studies will not change the risk estimate.", 3,
                                          np.where(d11["Q11c"] == "Irrespective of the quality of the study, future studies will change the risk estimate substantially anyway.", 4,0
                                                  )
                                          )
                                  )
                         )


d11["q8_4"] = np.where(d11["Q11d"] == "The growth rate will be 0.4% on average each year", 1,
                          np.where(d11["Q11d"] == "The growth rate over five years will be exactly 2%", 2,
                                  np.where(d11["Q11d"] == "The growth rate over five years will be between 1% and 3%", 3,
                                          np.where(d11["Q11d"] == "It is not possible to predict the growth rate with certainty", 4,0
                                                  )
                                          )
                                  )
                         )

d11["q8_5berlin_1"] = np.where(d11["Q11a_1"] == 1, 1, 2)
d11["q8_5london_1"] = np.where(d11["Q11a_1"] == 1, 1, 2)
d11["q8_5paris_1"] = np.where(d11["Q11a_1"] == 1, 1, 2)

d11["q8_6"] = np.where(d11["Q11i"] == "The medication increases recovery by 100%", 1,
                          np.where(d11["Q11i"] == "The medication increases recovery by 50%", 2,
                                   np.where(d11["Q11i"] == "The medication increases recovery by 2%", 3,
                                            np.where(d11["Q11i"] == "None of the above is implied", 4, 0
                                                    )
                                           )
                                  )
                         )

d11["q8_7"] = np.where(d11["Q11h"] == "5 in 100 people prior to the intervention to 6 out of 100 people after the intervention", 1,
                          np.where(d11["Q11h"] == "100 in 10,000 people prior to the intervention to 120 out of 10,000 people after the intervention", 2,
                                   np.where(d11["Q11h"] == "70 in 100 people prior to the intervention to 90 out of 100 people after the intervention", 3,
                                            np.where(d11["Q11h"] == "It is not possible to determine which of the answers is correct given the information provided", 4, 0
                                                    )
                                           )
                                  )
                         )


d11["q9_1_1"] = d11["Q14a"].copy()
d11["q9_2_1"] = d11["Q14b"].copy()

d11["q9_3"] = np.where(d11["Q14c"] == "Crosicol", 1,
                          np.where(d11["Q14c"] == "Hertinol", 2,
                                   np.where(d11["Q14c"] == "They are equal", 3,
                                            np.where(d11["Q14c"] == "Can’t say", 4, 0
                                                    )
                                           )
                                  )
                         )

d11["q10_1_1"] = d11["Q13a"].copy()
d11["q10_2_1"] = d11["Q13b"].copy()
d11["q10_3_1"] = d11["Q13c"].copy()

d11["q10_4"] = np.where(d11["Q13d"] == "9 out of 59", 1,
                          np.where(d11["Q13d"] == "9 out of 10", 2,
                                   np.where(d11["Q13d"] == "59 out of 1000", 3,
                                            np.where(d11["Q13d"] == "59 out of 100", 4, 0
                                                    )
                                           )
                                  )
                         )

d11[ matchCol ]

scoreColumns = ["certainty1", "certainty2", "certainty3", "certainty4", "certainty5", "uncertainty1", "uncertainty2", "numeracy1", "numeracy2", "numeracy3", "numeracy4", "numeracy5", "graph1", "graph2", "graph3", "riskcomprehension1", "riskcomprehension2", "riskcomprehension3", "riskcomprehension4", "riskcomprehension5", "bayesian1"]
d11[scoreColumns] = 0

d11Facet = d11.groupby(["ResponseId"]).progress_apply(scoring1)

100%|███████████████████████████████████████████████████████████████████████████████| 350/350 [00:02<00:00, 155.06it/s]

d11Facet

d11 = d11Facet.copy().reset_index(drop = True)

d12 = scoring2(d11)
d12

df2 = pd.concat([d01, d12], axis = 0).reset_index(drop = True)

df2

# The following graph represents frequency of each data point on "TotalScore_18" or the total score out of 13 questions across the sample.

df2['TotalScore_18'].plot(kind = 'hist', xticks = np.arange(0, 20, step=1), xlabel = 'TotalScore_18', title = 'Frequency of Scores' )

<Axes: title={'center': 'Frequency of Scores'}, xlabel='TotalScore_18', ylabel='Frequency'>

# Absolute mean scores for each facet

(df2[['Certainty_5', 'Uncertainty_2', 'RiskComprehension_5','GraphLiteracy_3','Numeracy_2','Bayesianreasoning_1','TotalScore_18']].mean(axis = 0)).plot(kind = 'bar', title = 'Absolute mean of scores for above data set')

<Axes: title={'center': 'Absolute mean of scores for above data set'}>

# Normalised mean scores for each facet

df2[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','GraphLiteracy_%','Numeracy_%','Bayesianreasoning_%','TotalScore_%']].mean(axis = 0).plot(kind = 'bar', title = '% mean of scores for above data set')

<Axes: title={'center': '% mean of scores for above data set'}>

# Data Frame 5 or df5 is an aggregate data on facet total scores and over all total score, along with  education data vs each response. 
# This data set is now sorted by education.

df5 = df2.sort_values(by = 'isced')
df5 = df5.reset_index(drop = True)

df5

# Mean of each facet as a % for each category of isced

df5.groupby('isced')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%']].mean()

# Count of responses for each category isced

df5.groupby('isced')[['ResponseId']].count()

# Absolute Total Facet scores mean line plot from low to high ISCED

df5.groupby('isced')[['TotalScore_18']].mean().plot( kind = 'line', title = 'Absolute Total Facet scores mean line plot', xticks = np.arange(1,4, step = 1)).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e7e8b7d0>

# Absolute Facet scores mean line plot from low to high edu

df5.groupby('isced')[['Certainty_5', "Uncertainty_2", 'RiskComprehension_5','Numeracy_2','GraphLiteracy_3','Bayesianreasoning_1']].mean().plot( kind = 'line', title = 'Absolute Facet scores mean line plot', xticks = np.arange(1,4, step = 1)).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e7e58290>

# Normalised Facet scores mean line plot from low to high edu


df5.groupby('isced')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%','TotalScore_%']].mean().plot( kind = 'line', title = 'Normalised Facet scores mean line plot', xticks = np.arange(1,4, step = 1)).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e61a7650>

# Normalised mean scores for each facet stacked

df5.groupby('isced')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%']].mean().plot( kind = 'bar', title = 'Normalised mean scores for each facet stacked', stacked = True).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e5f4d390>

# Normalised mean scores for each facet for each edu response category

df5.groupby('isced')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%','TotalScore_%']].mean().T.plot(kind = 'bar', title = 'Normalised mean scores for each facet for each edu response category').legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e5e43110>

# Trend line for Absolute Total Facet Score vs edu (isced) reponses

sns.regplot (data = df5, x = 'isced', y = 'TotalScore_18')

<Axes: xlabel='isced', ylabel='TotalScore_18'>

# Trend line for Absolute Independent Facet Score vs edu (isced) reponses

fig, ax6 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df5, x = 'isced', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax6, label='Certainty_5')
sns.regplot (data = df5, x = 'isced', y = 'Uncertainty_2', fit_reg=True, ci=None, ax=ax6, label='Uncertainty_2')
sns.regplot (data = df5, x = 'isced', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax6, label='RiskComprehension_5')
sns.regplot (data = df5, x = 'isced', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax6, label='Numeracy_2')
sns.regplot (data = df5, x = 'isced', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax6, label='GraphLiteracy_3')
sns.regplot (data = df5, x = 'isced', y = 'Bayesianreasoning_1',fit_reg=True, ci=None, ax=ax6, label='Bayesianreasoning_1' )

ax6.set(ylabel='Scores', xlabel='isced')
ax6.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Violine Plot for TotalScore_18 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'TotalScore_18')

<Axes: xlabel='isced', ylabel='TotalScore_18'>

# Violine Plot for Certainty_5 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'Certainty_5')

<Axes: xlabel='isced', ylabel='Certainty_5'>

# Violine Plot for Uncertainty_5 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'Uncertainty_2')

<Axes: xlabel='isced', ylabel='Uncertainty_2'>

# Violine Plot for RiskComprehension_5 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'RiskComprehension_5')

<Axes: xlabel='isced', ylabel='RiskComprehension_5'>

# Violine Plot for GraphLiteracy_3 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'GraphLiteracy_3')

<Axes: xlabel='isced', ylabel='GraphLiteracy_3'>

# Violine Plot for Numeracy_2 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'Numeracy_2')

<Axes: xlabel='isced', ylabel='Numeracy_2'>

# Violine Plot for TotalScore_18 for each entry under edu.

sns.violinplot( data = df5, x = 'isced', y = 'Bayesianreasoning_1')

<Axes: xlabel='isced', ylabel='Bayesianreasoning_1'>

# DF3 = Sorted by income

df3 = df2.sort_values(by = 'income')
df3 = df3.reset_index(drop = True)
df3.drop(df3[df3['income'] == 7].index, inplace = True)

df3

# Normalised mean of each facet as a numerical (sorted by income)

df3.groupby('income')[['Certainty_%','RiskComprehension_%','GraphLiteracy_%','Numeracy_%','Bayesianreasoning_%','TotalScore_%']].mean()

# Count of responses for each category (sorted by income)

df3.groupby('income')[['ResponseId']].count()

# Absolute mean of Total Facet score line plot (sorted by income)

df3.groupby('income')[['TotalScore_18']].mean().plot( kind = 'line', title = 'Absolute mean of Total Facet score line plot (sorted by income)').legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e7aead90>

# Absolute mean of each Facet score line plot (sorted by income)

df3.groupby('income')[['Certainty_5', "Uncertainty_2", 'RiskComprehension_5','GraphLiteracy_3','Numeracy_2','Bayesianreasoning_1']].mean().plot( kind = 'line', title = 'Absolute Facet scores mean line plot').legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e7b15b10>

# Normalised mean of each Facet score line plot (sorted by income)


df3.groupby('income')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','GraphLiteracy_%','Numeracy_%','Bayesianreasoning_%', 'TotalScore_%']].mean().plot( kind = 'line', title = '% mean of each Facet score line plot (sorted by income)').legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e79f3810>

# Normalised mean of each Facet score stacked plot (sorted by income)

df3.groupby('income')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','GraphLiteracy_%','Numeracy_%','Bayesianreasoning_%']].mean().plot( kind = 'bar', title = '% mean of each Facet score stacked plot (sorted by income)', stacked = True).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e7a42d50>

# Normalised mean of each Facet score hist plot (sorted by income)

df3.groupby('income')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','GraphLiteracy_%','Numeracy_%','Bayesianreasoning_%','TotalScore_%']].mean().T.plot(kind = 'bar', title = '% mean scores for each facet for each income response category').legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e731bb90>

# Trend line for Absolute Total Facet Score vs income

sns.regplot (data = df3, x = 'income', y = 'TotalScore_18')

<Axes: xlabel='income', ylabel='TotalScore_18'>

# Trend line for Absolute Independent Facet Score vs income reponses

fig, ax = plt.subplots(figsize=(6, 6))

sns.regplot (data = df3, x = 'income', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax, label='Certainty_5')
sns.regplot (data = df3, x = 'income', y = 'Uncertainty_2', fit_reg=True, ci=None, ax=ax, label='Uncertainty_2')
sns.regplot (data = df3, x = 'income', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax, label='RiskComprehension_5')
sns.regplot (data = df3, x = 'income', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax, label='Numeracy_2')
sns.regplot (data = df3, x = 'income', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax, label='GraphLiteracy_3')
sns.regplot (data = df3, x = 'income', y = 'Bayesianreasoning_1',fit_reg=True, ci=None, ax=ax, label='Bayesianreasoning_1' )

ax.set(ylabel='Scores', xlabel='income')
ax.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Violin Plot for TotalScore_18 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'TotalScore_18')

<Axes: xlabel='income', ylabel='TotalScore_18'>

# Violin Plot for Certainty_5 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'Certainty_5')

<Axes: xlabel='income', ylabel='Certainty_5'>

# Violin Plot for Certainty_5 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'Uncertainty_2')

<Axes: xlabel='income', ylabel='Uncertainty_2'>

# Violine Plot for RiskComprehension_5 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'RiskComprehension_5')

<Axes: xlabel='income', ylabel='RiskComprehension_5'>

# Violine Plot for GraphLiteracy_3 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'GraphLiteracy_3')

<Axes: xlabel='income', ylabel='GraphLiteracy_3'>

# Violine Plot for Numeracy_2 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'Numeracy_2')

<Axes: xlabel='income', ylabel='Numeracy_2'>

# Violine Plot for TotalScore_18 for each entry under income.

sns.violinplot( data = df3, x = 'income', y = 'Bayesianreasoning_1')

<Axes: xlabel='income', ylabel='Bayesianreasoning_1'>

# Descriptive stats for the data set, isced = 1
# NA values of income are removed

df7 = df5
df7.drop(df7[df7['income'] == 7].index, inplace = True)

df7.loc[df7['isced']==1].describe()

df7.loc[df7['isced']==2].describe()

df7.loc[df7['isced']==3].describe()

# Trend line for Absolute Tota Facet Score vs income reponses sorted by isced and ORDERED by income WITH scatter

fig, ax7 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = df7.loc[df7['isced']==1]['TotalScore_18'], fit_reg=True, ci=None, ax=ax7, label='ISCED = 1')
sns.regplot (data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = df7.loc[df7['isced']==2]['TotalScore_18'], fit_reg=True, ci=None, ax=ax7, label='ISCED = 2')
sns.regplot (data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = df7.loc[df7['isced']==3]['TotalScore_18'], fit_reg=True, ci=None, ax=ax7, label='ISCED = 3')

ax7.set(ylabel='Total Scores_19', xlabel='INCOME')
ax7.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Tota Facet Score vs income reponses sorted by isced and ORDERED by income WITHOUT scatter

fig, ax8 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 3')

ax8.set(ylabel='Total Scores_19', xlabel='INCOME')
ax8.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Certainty Score vs income reponses sorted by isced and ORDERED by income WITHOUT scatter

fig, ax9 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 3')

ax9.set(ylabel='Certainty_5', xlabel='INCOME')
ax9.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Risk Comprehension Score vs income reponses sorted by isced and ORDERED by income WITHOUT scatter

fig, ax10 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 3')

ax10.set(ylabel='RiskComprehension_5', xlabel='INCOME')
ax10.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Number Comprehension Score vs income reponses sorted by isced and ORDERED by income WITHOUT scatter

fig, ax11 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 3')

ax11.set(ylabel='Numeracy_2', xlabel='INCOME')
ax11.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Graph Comprehension Score vs income reponses sorted by isced and ORDERED by income WITHOUT scatter

fig, ax11 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 3')

ax11.set(ylabel='GraphLiteracy_3', xlabel='INCOME')
ax11.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Bayesian Reasoning Score vs income reponses sorted by isced and ORDERED by income WITHOUT scatter

fig, ax12 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[df7['isced']==1].reset_index(drop = True), x = 'income', y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==2].reset_index(drop = True), x = 'income', y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[df7['isced']==3].reset_index(drop = True), x = 'income', y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 3')

ax12.set(ylabel='Bayesianreasoning_1', xlabel='INCOME')
ax12.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

df7

# Since we already have a classification for Age groups in the form of age_rec, we will use that.
# We will also use median of age to see if it yields any relevant results, as instructed.

# AXES to be used = Age or age groups, ISCED, Income

# Age groups = [2,3,4]
# Age group 2 = 18 to 35 y/o
# Age group 3 = 36 to 59 y/o
# Age group 4 = 60 to 75 y/o (75 y/o, i.e, within the scope of the data we have, it can mean 60 and above also)

df7.loc[(df7['age_rec']==2)].describe()

df7.loc[(df7['age_rec']==3)].describe()

df7.loc[(df7['age_rec']==4)].describe()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Total Scores_14', xlabel='INCOME for age 18 to 35')
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Total Scores_14', xlabel='INCOME for age 36 to 55')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Total Scores_14', xlabel='INCOME for age 56 and above')
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ df7['age_rec']==2  ]["income"].unique(), y = df7.loc[ df7['age_rec']==2  ].groupby(['income']).mean(numeric_only=True)['TotalScore_18'], yerr = df7.loc[ (df7['age_rec']==2) ].groupby(['income'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Total Scores_14', xlabel='INCOME for age 18 to 35',  yticks = np.arange(5, 16 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Total Scores_14', xlabel='INCOME for age 36 to 55',  yticks = np.arange(5, 16 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df7.loc[ df7['age_rec']==3  ]["income"].unique(), y = df7.loc[ df7['age_rec']==3  ].groupby(['income']).mean(numeric_only=True)['TotalScore_18'], yerr = df7.loc[ (df7['age_rec']==3) ].groupby(['income'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Total Scores_14', xlabel='INCOME for age 56 and above',  yticks = np.arange(5, 16 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df7.loc[ df7['age_rec']== 4 ]['income'].unique(), y = df7.loc[ df7['age_rec']==4  ].groupby(['income']).mean(numeric_only=True)['TotalScore_18'], yerr = df7.loc[ (df7['age_rec']==4) ].groupby(['income'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Certainty_5', xlabel='INCOME for age 18 to 35',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df7.loc[ df7['age_rec']==2  ]["income"].unique(), y = df7.loc[ df7['age_rec']==2  ].groupby(['income'])['Certainty_5'].mean(), yerr = df7.loc[ (df7['age_rec']==2) ].groupby(['income'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Certainty_5', xlabel='INCOME for age 36 to 55',  yticks = np.arange(0, 6 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df7.loc[ df7['age_rec']==3  ]["income"].unique(), y = df7.loc[ df7['age_rec']==3  ].groupby(['income'])['Certainty_5'].mean(), yerr = df7.loc[ (df7['age_rec']==3) ].groupby(['income'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Certainty_5', xlabel='INCOME for age 56 and above',  yticks = np.arange(0, 6 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df7.loc[ df7['age_rec']== 4 ]['income'].unique(), y = df7.loc[ df7['age_rec']==4  ].groupby(['income'])['Certainty_5'].mean(), yerr = df7.loc[ (df7['age_rec']==4) ].groupby(['income'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='RiskComprehension_5', xlabel='INCOME for age 18 to 35',  yticks = np.arange(0, 3 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df7.loc[ df7['age_rec']==2  ]["income"].unique(), y = df7.loc[ df7['age_rec']==2  ].groupby(['income'])['RiskComprehension_5'].mean(), yerr = df7.loc[ (df7['age_rec']==2) ].groupby(['income'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='RiskComprehension_5', xlabel='INCOME for age 36 to 55',  yticks = np.arange(0, 3 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df7.loc[ df7['age_rec']==3  ]["income"].unique(), y = df7.loc[ df7['age_rec']==3  ].groupby(['income'])['RiskComprehension_5'].mean(), yerr = df7.loc[ (df7['age_rec']==3) ].groupby(['income'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='RiskComprehension_5', xlabel='INCOME for age 56 and above',  yticks = np.arange(0, 3 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df7.loc[ df7['age_rec']== 4 ]['income'].unique(), y = df7.loc[ df7['age_rec']==4  ].groupby(['income'])['RiskComprehension_5'].mean(), yerr = df7.loc[ (df7['age_rec']==4) ].groupby(['income'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Numeracy_2', xlabel='INCOME for age 18 to 35',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df7.loc[ df7['age_rec']==2  ]["income"].unique(), y = df7.loc[ df7['age_rec']==2  ].groupby(['income'])['Numeracy_2'].mean(), yerr = df7.loc[ (df7['age_rec']==2) ].groupby(['income'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Numeracy_2', xlabel='INCOME for age 36 to 55',  yticks = np.arange(0, 6 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df7.loc[ df7['age_rec']==3  ]["income"].unique(), y = df7.loc[ df7['age_rec']==3  ].groupby(['income'])['Numeracy_2'].mean(), yerr = df7.loc[ (df7['age_rec']==3) ].groupby(['income'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Numeracy_2', xlabel='INCOME for age 56 and above',  yticks = np.arange(0, 6 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df7.loc[ df7['age_rec']== 4 ]['income'].unique(), y = df7.loc[ df7['age_rec']==4  ].groupby(['income'])['Numeracy_2'].mean(), yerr = df7.loc[ (df7['age_rec']==4) ].groupby(['income'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='GraphLiteracy_3', xlabel='INCOME for age 18 to 35',  yticks = np.arange(0, 4 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df7.loc[ df7['age_rec']==2  ]["income"].unique(), y = df7.loc[ df7['age_rec']==2  ].groupby(['income'])['GraphLiteracy_3'].mean(), yerr = df7.loc[ (df7['age_rec']==2) ].groupby(['income'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='GraphLiteracy_3', xlabel='INCOME for age 36 to 55',  yticks = np.arange(0, 4 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df7.loc[ df7['age_rec']==3  ]["income"].unique(), y = df7.loc[ df7['age_rec']==3  ].groupby(['income'])['GraphLiteracy_3'].mean(), yerr = df7.loc[ (df7['age_rec']==3) ].groupby(['income'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='GraphLiteracy_3', xlabel='INCOME for age 56 and above',  yticks = np.arange(0, 4 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df7.loc[ df7['age_rec']== 4 ]['income'].unique(), y = df7.loc[ df7['age_rec']==4  ].groupby(['income'])['GraphLiteracy_3'].mean(), yerr = df7.loc[ (df7['age_rec']==4) ].groupby(['income'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==2) & (df7['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Bayesianreasoning_1', xlabel='INCOME for age 18 to 35',  yticks = np.arange(0, 5 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df7.loc[ df7['age_rec']==2  ]["income"].unique(), y = df7.loc[ df7['age_rec']==2  ].groupby(['income'])['Bayesianreasoning_1'].mean(), yerr = df7.loc[ (df7['age_rec']==2) ].groupby(['income'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==3) & (df7['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Bayesianreasoning_1', xlabel='INCOME for age 36 to ',  yticks = np.arange(0, 5 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df7.loc[ df7['age_rec']==3  ]["income"].unique(), y = df7.loc[ df7['age_rec']==3  ].groupby(['income'])['Bayesianreasoning_1'].mean(), yerr = df7.loc[ (df7['age_rec']==3) ].groupby(['income'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age_rec']==4) & (df7['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Bayesianreasoning_1', xlabel='INCOME for age 56 and above',  yticks = np.arange(0, 5 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df7.loc[ df7['age_rec']== 4 ]['income'].unique(), y = df7.loc[ df7['age_rec']==4  ].groupby(['income'])['Bayesianreasoning_1'].mean(), yerr = df7.loc[ (df7['age_rec']==4) ].groupby(['income'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

# Age mean method
# Age_mean

age_m = df2["age"].median()
age_m

30.5

df7["income"].value_counts()

income
2    329
1    298
3    110
4     65
5     37
Name: count, dtype: int64

df7.loc[(df7['age'] < age_m)].describe()

df7.loc[(df7['age'] >= age_m)].describe()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Total Scores_19', xlabel='INCOME for < median age')
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df7.loc[(df7['age'] >= age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Total Scores_19', xlabel='INCOME for >=median age')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ (df7['age'] < age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] < age_m)  ].groupby(['income'])['TotalScore_18'].mean(), yerr = df7.loc[ (df7['age'] < age_m) ].groupby(['income'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Total Scores_19', xlabel='INCOME for < median age',  yticks = np.arange(5, 16 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df7.loc[ (df7['age'] >= age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] >=age_m)  ].groupby(['income'])['TotalScore_18'].mean(), yerr = df7.loc[ (df7['age'] >=age_m) ].groupby(['income'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Total Scores_19', xlabel='INCOME for >=median age',  yticks = np.arange(5, 16 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ (df7['age'] < age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] < age_m)  ].groupby(['income'])['Certainty_5'].mean(), yerr = df7.loc[ (df7['age'] < age_m) ].groupby(['income'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Certainty_5', xlabel='INCOME for < median age',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df7.loc[ (df7['age'] >= age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] >=age_m)  ].groupby(['income'])['Certainty_5'].mean(), yerr = df7.loc[ (df7['age'] >=age_m) ].groupby(['income'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Certainty_5', xlabel='INCOME for >=median age')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ (df7['age'] < age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] < age_m)  ].groupby(['income'])['RiskComprehension_5'].mean(), yerr = df7.loc[ (df7['age'] < age_m) ].groupby(['income'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='RiskComprehension_5', xlabel='INCOME for < median age',  yticks = np.arange(0, 3 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df7.loc[ (df7['age'] >= age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] >=age_m)  ].groupby(['income'])['RiskComprehension_5'].mean(), yerr = df7.loc[ (df7['age'] >=age_m) ].groupby(['income'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='RiskComprehension_5', xlabel='INCOME for >=median age',  yticks = np.arange(0, 3 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ (df7['age'] < age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] < age_m)  ].groupby(['income'])['Numeracy_2'].mean(), yerr = df7.loc[ (df7['age'] < age_m) ].groupby(['income'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Numeracy_2', xlabel='INCOME for < median age',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df7.loc[ (df7['age'] >= age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] >=age_m)  ].groupby(['income'])['Numeracy_2'].mean(), yerr = df7.loc[ (df7['age'] >=age_m) ].groupby(['income'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Numeracy_2', xlabel='INCOME for >=median age',  yticks = np.arange(0, 6 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ (df7['age'] < age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] < age_m)  ].groupby(['income'])['GraphLiteracy_3'].mean(), yerr = df7.loc[ (df7['age'] < age_m) ].groupby(['income'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='GraphLiteracy_3', xlabel='INCOME for < median age',  yticks = np.arange(0, 4 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df7.loc[ (df7['age'] >= age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] >=age_m)  ].groupby(['income'])['GraphLiteracy_3'].mean(), yerr = df7.loc[ (df7['age'] >=age_m) ].groupby(['income'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='GraphLiteracy_3', xlabel='INCOME for >=median age',  yticks = np.arange(0, 4 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] < age_m) & (df7['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df7.loc[ (df7['age'] < age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] < age_m)  ].groupby(['income'])['Bayesianreasoning_1'].mean(), yerr = df7.loc[ (df7['age'] < age_m) ].groupby(['income'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Bayesianreasoning_1', xlabel='INCOME for < median age',  yticks = np.arange(0, 5 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)].reset_index(drop = True), x = 'income', y = df7.loc[(df7['age'] >=age_m) & (df7['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df7.loc[ (df7['age'] >= age_m)  ]["income"].unique(), y = df7.loc[ (df7['age'] >=age_m)  ].groupby(['income'])['Bayesianreasoning_1'].mean(), yerr = df7.loc[ (df7['age'] >=age_m) ].groupby(['income'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Bayesianreasoning_1', xlabel='INCOME for >=median age',  yticks = np.arange(0, 5 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Data Frame 2 or df2 is an aggregate data on facet total scores and over all total score, along with wealth and income data vs each response. 
# This data set is also sorted by wealth with all responses 'NA' removed.
# We also assign quartiles ranking each response according to this sort.

df4 = df2.sort_values(by = 'wealth')
df4 = df4.reset_index(drop = True)
df4 = df4.dropna(axis = 0, subset = 'wealth')
df4.insert(loc = len(df4.columns), column = "Quartile Number", value = pd.qcut(df4["wealth"],q = 4, labels = False ) + 1, allow_duplicates = 'False')

df4

# Trend line for Absolute Total Facet Score vs unique wealth reponses

sns.regplot (data = df4, x = df4.index, y = 'TotalScore_18')

<Axes: ylabel='TotalScore_18'>

# Trend line for Absolute Independent Facet Score vs wealth reponses

fig1, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4, x = df4.index, y = 'Certainty_5', fit_reg=True, ci=None, ax=ax1, label='Certainty_5')
sns.regplot (data = df4, x = df4.index, y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax1, label='RiskComprehension_5')
sns.regplot (data = df4, x = df4.index, y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax1, label='Numeracy_2')
sns.regplot (data = df4, x = df4.index, y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax1, label='GraphLiteracy_3')
sns.regplot (data = df4, x = df4.index, y = 'Bayesianreasoning_1',fit_reg=True, ci=None, ax=ax1, label='Bayesianreasoning_1' )

ax1.set(ylabel='Scores', xlabel='wealth')
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Count of number of responses under each quartile

df4.groupby('Quartile Number')[['Quartile Number']].count()

# Absolute Total Facet scores mean line plot from low to high Wealth

df4.groupby('Quartile Number')[['TotalScore_18']].mean().plot( kind = 'line', title = 'Absolute Total Facet scores mean line plot low to high wealth', xticks = np.arange(1,5,step = 1)).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e6086410>

# Absolute Facet scores mean line plot from low to high wealth

df4.groupby('Quartile Number')[['Certainty_5', "Uncertainty_2", 'RiskComprehension_5','Numeracy_2','GraphLiteracy_3','Bayesianreasoning_1']].mean().plot( kind = 'line', title = 'Absolute Facet scores mean line plot low to high wealth', xticks = np.arange(1,5,step = 1)).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e66c8c90>

# Normalised Facet scores mean line plot from low to high wealth


df4.groupby('Quartile Number')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%','TotalScore_%']].mean().plot( kind = 'line', title = '% Facet scores mean line plot from low to high wealth',  xticks = np.arange(1,5,step = 1)).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e749cb50>

# Normalised mean scores for each facet stacked

df4.groupby('Quartile Number')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%']].mean().plot( kind = 'bar', title = '% mean scores for each facet stacked', stacked = True).legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293ec890ad0>

# Normalised mean scores for each facet for each wealth response Quartile

df4.groupby('Quartile Number')[['Certainty_%', "Uncertainty_%", 'RiskComprehension_%','Numeracy_%','GraphLiteracy_%','Bayesianreasoning_%','TotalScore_%']].mean().T.plot(kind = 'bar', title = '% mean scores for each facet for each wealth response category').legend(loc='center left',bbox_to_anchor=(1.0, 0.5))

<matplotlib.legend.Legend at 0x293e73a59d0>

# Trend line for Absolute Total Facet Score vs Quartiles

sns.regplot (data = df4, x = 'Quartile Number', y = 'TotalScore_18')

<Axes: xlabel='Quartile Number', ylabel='TotalScore_18'>

# Trend line for Absolute Independent Facet Score vs Quartiles

fig2, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4, x = 'Quartile Number', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax2, label='Certainty_5')
sns.regplot (data = df4, x = 'Quartile Number', y = 'Uncertainty_2', fit_reg=True, ci=None, ax=ax2, label='Uncertainty_2')
sns.regplot (data = df4, x = 'Quartile Number', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax2, label='RiskComprehension_5')
sns.regplot (data = df4, x = 'Quartile Number', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax2, label='Numeracy_2')
sns.regplot (data = df4, x = 'Quartile Number', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax2, label='GraphLiteracy_3')
sns.regplot (data = df4, x = 'Quartile Number', y = 'Bayesianreasoning_1',fit_reg=True, ci=None, ax=ax2, label='Bayesianreasoning_1' )

ax2.set(ylabel='Scores', xlabel='wealth')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Violine Plot for TotalScore_18 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'TotalScore_18')

<Axes: xlabel='Quartile Number', ylabel='TotalScore_18'>

# Violine Plot for Certainty_5 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'Certainty_5')

<Axes: xlabel='Quartile Number', ylabel='Certainty_5'>

# Violine Plot for Certainty_5 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'Uncertainty_2')

<Axes: xlabel='Quartile Number', ylabel='Uncertainty_2'>

# Violine Plot for RiskComprehension_5 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'RiskComprehension_5')

<Axes: xlabel='Quartile Number', ylabel='RiskComprehension_5'>

# Violine Plot for GraphLiteracy_3 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'GraphLiteracy_3')

<Axes: xlabel='Quartile Number', ylabel='GraphLiteracy_3'>

# Violine Plot for Numeracy_2 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'Numeracy_2')

<Axes: xlabel='Quartile Number', ylabel='Numeracy_2'>

# Violine Plot for Bayesianreasoning_1 for each entry under Wealth Quartiles.

sns.violinplot( data = df4, x = 'Quartile Number', y = 'Bayesianreasoning_1')

<Axes: xlabel='Quartile Number', ylabel='Bayesianreasoning_1'>

# Descriptive stats for the data set, isced = 1
# NA values of wealth are removed

df6 = df5.dropna(axis = 0, subset = 'wealth')
# df6.drop(df6[df6['wealth'] == 3500000].index, inplace = True)
df6.loc[df6['isced']==1][['wealth']].describe()

# Descriptive stats for the data set, isced = 2

df6.loc[df6['isced']==2][['wealth']].describe()

# Descriptive stats for the data set, isced = 3

df6.loc[df6['isced']==3][['wealth']].describe()

# Trend line for Absolute Tota Facet Score vs wealth reponses sorted by isced and ORDERED by wealth WITH scatter

fig, ax7 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax7, label='ISCED = 1')
sns.regplot (data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax7, label='ISCED = 2')
sns.regplot (data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax7, label='ISCED = 3')

ax7.set(ylabel='Total Scores_19', xlabel='Wealth')
ax7.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Tota Facet Score vs wealth reponses sorted by isced and ORDERED by wealth WITHOUT scatter

fig, ax8 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 1')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 2')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 3')

ax8.set(ylabel='Total Scores_19', xlabel='Wealth')
ax8.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Certainty Score vs wealth reponses sorted by isced and ORDERED by wealth WITHOUT scatter

fig, ax9 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 1')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 2')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 3')

ax9.set(ylabel='Certainty_5', xlabel='Wealth')
ax9.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Uncertainty Score vs wealth reponses sorted by isced and ORDERED by wealth WITHOUT scatter

fig, ax10 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 1')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 2')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 3')

ax10.set(ylabel='RiskComprehension_5', xlabel='Wealth')
ax10.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Number Comprehension Score vs wealth reponses sorted by isced and ORDERED by wealth WITHOUT scatter

fig, ax11 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 1')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 2')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 3')

ax11.set(ylabel='Numeracy_2', xlabel='Wealth')
ax11.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Graph Comprehension Score vs wealth reponses sorted by isced and ORDERED by wealth WITHOUT scatter

fig, ax11 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 1')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 2')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 3')

ax11.set(ylabel='GraphLiteracy_3', xlabel='Wealth')
ax11.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Bayesian Reasoning Score vs wealth reponses sorted by isced and ORDERED by wealth WITHOUT scatter

fig, ax12 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df6.loc[df6['isced']==1].reset_index(drop = True), x = df6.loc[df6['isced']==1].reset_index(drop = True)['wealth'], y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 1')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==2].reset_index(drop = True), x = df6.loc[df6['isced']==2].reset_index(drop = True)['wealth'], y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 2')
sns.regplot (scatter = False, data = df6.loc[df6['isced']==3].reset_index(drop = True), x = df6.loc[df6['isced']==3].reset_index(drop = True)['wealth'], y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 3')

ax12.set(ylabel='Bayesianreasoning_1', xlabel='Wealth')
ax12.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Descriptive stats for the data set, isced = 1
# NA values of wealth are removed

df4.loc[df4['isced']==1].describe()

df4.loc[df4['isced']==2].describe()

df4.loc[df4['isced']==3].describe()

# Trend line for Absolute Tota Facet Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITH scatter

fig, ax7 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[df4['isced']==1]['TotalScore_18'], fit_reg=True, ci=None, ax=ax7, label='ISCED = 1')
sns.regplot (data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[df4['isced']==2]['TotalScore_18'], fit_reg=True, ci=None, ax=ax7, label='ISCED = 2')
sns.regplot (data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[df4['isced']==3]['TotalScore_18'], fit_reg=True, ci=None, ax=ax7, label='ISCED = 3')

ax7.set(ylabel='Total Scores_19', xlabel='WEALTH Quartile Number')
ax7.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Tota Facet Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITHOUT scatter

fig, ax8 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = 'TotalScore_18', fit_reg=True, ci=None, ax=ax8, label='ISCED = 3')

ax8.set(ylabel='Total Scores_19', xlabel='WEALTH Quartile Number')
ax8.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Certainty Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITHOUT scatter

fig, ax9 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = 'Certainty_5', fit_reg=True, ci=None, ax=ax9, label='ISCED = 3')

ax9.set(ylabel='Certainty_5', xlabel='WEALTH Quartile Number')
ax9.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Risk Comprehension Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITHOUT scatter

fig, ax10 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = 'RiskComprehension_5', fit_reg=True, ci=None, ax=ax10, label='ISCED = 3')

ax10.set(ylabel='RiskComprehension_5', xlabel='WEALTH Quartile Number')
ax10.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Number Comprehension Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITHOUT scatter

fig, ax11 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = 'Numeracy_2', fit_reg=True, ci=None, ax=ax11, label='ISCED = 3')

ax11.set(ylabel='Numeracy_2', xlabel='WEALTH Quartile Number')
ax11.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Graph Comprehension Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITHOUT scatter

fig, ax11 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = 'GraphLiteracy_3', fit_reg=True, ci=None, ax=ax11, label='ISCED = 3')

ax11.set(ylabel='GraphLiteracy_3', xlabel='WEALTH Quartile Number')
ax11.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

# Trend line for Absolute Bayesian Reasoning Score vs Quartile Number reponses sorted by isced and ORDERED by Quartile Number WITHOUT scatter

fig, ax12 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[df4['isced']==1].reset_index(drop = True), x = 'Quartile Number', y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==2].reset_index(drop = True), x = 'Quartile Number', y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[df4['isced']==3].reset_index(drop = True), x = 'Quartile Number', y = 'Bayesianreasoning_1', fit_reg=True, ci=None, ax=ax12, label='ISCED = 3')

ax12.set(ylabel='Bayesianreasoning_1', xlabel='WEALTH Quartile Number')
ax12.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

df4

# Since we already have a classification for Age groups in the form of age_rec, we will use that.
# We will also use median of age to see if it yields any relevant results, as instructed.

# AXES to be used = Age or age groups, ISCED, Income

# Age groups = [2,3,4]
# Age group 2 = 18 to 35 y/o
# Age group 3 = 36 to 59 y/o
# Age group 4 = 60 to 75 y/o (75 y/o, i.e, within the scope of the data we have, it can mean 60 and above also)

df4.loc[(df4['age_rec']==2)].describe()

df4.loc[(df4['age_rec']==3)].describe()

df4.loc[(df4['age_rec']==4)].describe()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Total Scores_14', xlabel='WEALTH Quartile Number for age 18 to 35')
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Total Scores_14', xlabel='WEALTH Quartile Number for age 36 to 55')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Total Scores_14', xlabel='WEALTH Quartile Number for age 56 and above')
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ df4['age_rec']==2  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==2  ].groupby(['Quartile Number']).mean(numeric_only=True)['TotalScore_18'], yerr = df4.loc[ (df4['age_rec']==2) ].groupby(['Quartile Number'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Total Scores_14', xlabel='WEALTH Quartile Number for age 18 to 35',  yticks = np.arange(5, 16 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Total Scores_14', xlabel='WEALTH Quartile Number for age 36 to 55',  yticks = np.arange(5, 16 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df4.loc[ df4['age_rec']==3  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==3  ].groupby(['Quartile Number']).mean(numeric_only=True)['TotalScore_18'], yerr = df4.loc[ (df4['age_rec']==3) ].groupby(['Quartile Number'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Total Scores_14', xlabel='WEALTH Quartile Number for age 56 and above',  yticks = np.arange(5, 16 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df4.loc[ df4['age_rec']== 4 ]['Quartile Number'].unique(), y = df4.loc[ df4['age_rec']==4  ].groupby(['Quartile Number']).mean(numeric_only=True)['TotalScore_18'], yerr = df4.loc[ (df4['age_rec']==4) ].groupby(['Quartile Number'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Certainty_5', xlabel='WEALTH Quartile Number for age 18 to 35',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df4.loc[ df4['age_rec']==2  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==2  ].groupby(['Quartile Number'])['Certainty_5'].mean(), yerr = df4.loc[ (df4['age_rec']==2) ].groupby(['Quartile Number'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Certainty_5', xlabel='WEALTH Quartile Number for age 36 to 55',  yticks = np.arange(0, 6 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df4.loc[ df4['age_rec']==3  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==3  ].groupby(['Quartile Number'])['Certainty_5'].mean(), yerr = df4.loc[ (df4['age_rec']==3) ].groupby(['Quartile Number'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Certainty_5', xlabel='WEALTH Quartile Number for age 56 and above',  yticks = np.arange(0, 6 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df4.loc[ df4['age_rec']== 4 ]['Quartile Number'].unique(), y = df4.loc[ df4['age_rec']==4  ].groupby(['Quartile Number'])['Certainty_5'].mean(), yerr = df4.loc[ (df4['age_rec']==4) ].groupby(['Quartile Number'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='RiskComprehension_5', xlabel='WEALTH Quartile Number for age 18 to 35',  yticks = np.arange(0, 3 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df4.loc[ df4['age_rec']==2  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==2  ].groupby(['Quartile Number'])['RiskComprehension_5'].mean(), yerr = df4.loc[ (df4['age_rec']==2) ].groupby(['Quartile Number'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='RiskComprehension_5', xlabel='WEALTH Quartile Number for age 36 to 55',  yticks = np.arange(0, 3 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df4.loc[ df4['age_rec']==3  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==3  ].groupby(['Quartile Number'])['RiskComprehension_5'].mean(), yerr = df4.loc[ (df4['age_rec']==3) ].groupby(['Quartile Number'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='RiskComprehension_5', xlabel='WEALTH Quartile Number for age 56 and above',  yticks = np.arange(0, 3 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df4.loc[ df4['age_rec']== 4 ]['Quartile Number'].unique(), y = df4.loc[ df4['age_rec']==4  ].groupby(['Quartile Number'])['RiskComprehension_5'].mean(), yerr = df4.loc[ (df4['age_rec']==4) ].groupby(['Quartile Number'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Numeracy_2', xlabel='WEALTH Quartile Number for age 18 to 35',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df4.loc[ df4['age_rec']==2  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==2  ].groupby(['Quartile Number'])['Numeracy_2'].mean(), yerr = df4.loc[ (df4['age_rec']==2) ].groupby(['Quartile Number'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Numeracy_2', xlabel='WEALTH Quartile Number for age 36 to 55',  yticks = np.arange(0, 6 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df4.loc[ df4['age_rec']==3  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==3  ].groupby(['Quartile Number'])['Numeracy_2'].mean(), yerr = df4.loc[ (df4['age_rec']==3) ].groupby(['Quartile Number'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Numeracy_2', xlabel='WEALTH Quartile Number for age 56 and above',  yticks = np.arange(0, 6 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df4.loc[ df4['age_rec']== 4 ]['Quartile Number'].unique(), y = df4.loc[ df4['age_rec']==4  ].groupby(['Quartile Number'])['Numeracy_2'].mean(), yerr = df4.loc[ (df4['age_rec']==4) ].groupby(['Quartile Number'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='GraphLiteracy_3', xlabel='WEALTH Quartile Number for age 18 to 35',  yticks = np.arange(0, 4 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df4.loc[ df4['age_rec']==2  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==2  ].groupby(['Quartile Number'])['GraphLiteracy_3'].mean(), yerr = df4.loc[ (df4['age_rec']==2) ].groupby(['Quartile Number'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='GraphLiteracy_3', xlabel='WEALTH Quartile Number for age 36 to 55',  yticks = np.arange(0, 4 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df4.loc[ df4['age_rec']==3  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==3  ].groupby(['Quartile Number'])['GraphLiteracy_3'].mean(), yerr = df4.loc[ (df4['age_rec']==3) ].groupby(['Quartile Number'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='GraphLiteracy_3', xlabel='WEALTH Quartile Number for age 56 and above',  yticks = np.arange(0, 4 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df4.loc[ df4['age_rec']== 4 ]['Quartile Number'].unique(), y = df4.loc[ df4['age_rec']==4  ].groupby(['Quartile Number'])['GraphLiteracy_3'].mean(), yerr = df4.loc[ (df4['age_rec']==4) ].groupby(['Quartile Number'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==2) & (df4['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Bayesianreasoning_1', xlabel='WEALTH Quartile Number for age 18 to 35',  yticks = np.arange(0, 5 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax1.errorbar(x = df4.loc[ df4['age_rec']==2  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==2  ].groupby(['Quartile Number'])['Bayesianreasoning_1'].mean(), yerr = df4.loc[ (df4['age_rec']==2) ].groupby(['Quartile Number'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==3) & (df4['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Bayesianreasoning_1', xlabel='WEALTH Quartile Number for age 36 to ',  yticks = np.arange(0, 5 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax2.errorbar(x = df4.loc[ df4['age_rec']==3  ]["Quartile Number"].unique(), y = df4.loc[ df4['age_rec']==3  ].groupby(['Quartile Number'])['Bayesianreasoning_1'].mean(), yerr = df4.loc[ (df4['age_rec']==3) ].groupby(['Quartile Number'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

fig, ax3 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age_rec']==4) & (df4['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax3, label='ISCED = 3')

ax3.set(ylabel='Bayesianreasoning_1', xlabel='WEALTH Quartile Number for age 56 and above',  yticks = np.arange(0, 5 , 1))
ax3.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
ax3.errorbar(x = df4.loc[ df4['age_rec']== 4 ]['Quartile Number'].unique(), y = df4.loc[ df4['age_rec']==4  ].groupby(['Quartile Number'])['Bayesianreasoning_1'].mean(), yerr = df4.loc[ (df4['age_rec']==4) ].groupby(['Quartile Number'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
plt.show()

# Age mean method
# Age_mean

age_m = df2["age"].median()
age_m

30.5

df4["Quartile Number"].value_counts()

Quartile Number
2    252
1    221
3    195
4    171
Name: count, dtype: int64

df4.loc[(df4['age'] < age_m)].describe()

df4.loc[(df4['age'] >= age_m)].describe()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.set(ylabel='Total Scores_19', xlabel='WEALTH Quartile Number for < median age')
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (data = df4.loc[(df4['age'] >= age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.set(ylabel='Total Scores_19', xlabel='WEALTH Quartile Number for >=median age')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ (df4['age'] < age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] < age_m)  ].groupby(['Quartile Number'])['TotalScore_18'].mean(), yerr = df4.loc[ (df4['age'] < age_m) ].groupby(['Quartile Number'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Total Scores_19', xlabel='WEALTH Quartile Number for < median age',  yticks = np.arange(5, 16 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['TotalScore_18'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df4.loc[ (df4['age'] >= age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] >=age_m)  ].groupby(['Quartile Number'])['TotalScore_18'].mean(), yerr = df4.loc[ (df4['age'] >=age_m) ].groupby(['Quartile Number'])['TotalScore_18'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Total Scores_19', xlabel='WEALTH Quartile Number for >=median age',  yticks = np.arange(5, 16 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ (df4['age'] < age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] < age_m)  ].groupby(['Quartile Number'])['Certainty_5'].mean(), yerr = df4.loc[ (df4['age'] < age_m) ].groupby(['Quartile Number'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Certainty_5', xlabel='WEALTH Quartile Number for < median age',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['Certainty_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df4.loc[ (df4['age'] >= age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] >=age_m)  ].groupby(['Quartile Number'])['Certainty_5'].mean(), yerr = df4.loc[ (df4['age'] >=age_m) ].groupby(['Quartile Number'])['Certainty_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Certainty_5', xlabel='WEALTH Quartile Number for >=median age')
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ (df4['age'] < age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] < age_m)  ].groupby(['Quartile Number'])['RiskComprehension_5'].mean(), yerr = df4.loc[ (df4['age'] < age_m) ].groupby(['Quartile Number'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='RiskComprehension_5', xlabel='WEALTH Quartile Number for < median age',  yticks = np.arange(0, 3 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['RiskComprehension_5'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df4.loc[ (df4['age'] >= age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] >=age_m)  ].groupby(['Quartile Number'])['RiskComprehension_5'].mean(), yerr = df4.loc[ (df4['age'] >=age_m) ].groupby(['Quartile Number'])['RiskComprehension_5'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='RiskComprehension_5', xlabel='WEALTH Quartile Number for >=median age',  yticks = np.arange(0, 3 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ (df4['age'] < age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] < age_m)  ].groupby(['Quartile Number'])['Numeracy_2'].mean(), yerr = df4.loc[ (df4['age'] < age_m) ].groupby(['Quartile Number'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Numeracy_2', xlabel='WEALTH Quartile Number for < median age',  yticks = np.arange(0, 6 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['Numeracy_2'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df4.loc[ (df4['age'] >= age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] >=age_m)  ].groupby(['Quartile Number'])['Numeracy_2'].mean(), yerr = df4.loc[ (df4['age'] >=age_m) ].groupby(['Quartile Number'])['Numeracy_2'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Numeracy_2', xlabel='WEALTH Quartile Number for >=median age',  yticks = np.arange(0, 6 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ (df4['age'] < age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] < age_m)  ].groupby(['Quartile Number'])['GraphLiteracy_3'].mean(), yerr = df4.loc[ (df4['age'] < age_m) ].groupby(['Quartile Number'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='GraphLiteracy_3', xlabel='WEALTH Quartile Number for < median age',  yticks = np.arange(0, 4 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['GraphLiteracy_3'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df4.loc[ (df4['age'] >= age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] >=age_m)  ].groupby(['Quartile Number'])['GraphLiteracy_3'].mean(), yerr = df4.loc[ (df4['age'] >=age_m) ].groupby(['Quartile Number'])['GraphLiteracy_3'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='GraphLiteracy_3', xlabel='WEALTH Quartile Number for >=median age',  yticks = np.arange(0, 4 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax1 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] < age_m) & (df4['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax1, label='ISCED = 3')

ax1.errorbar(x = df4.loc[ (df4['age'] < age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] < age_m)  ].groupby(['Quartile Number'])['Bayesianreasoning_1'].mean(), yerr = df4.loc[ (df4['age'] < age_m) ].groupby(['Quartile Number'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax1.set(ylabel='Bayesianreasoning_1', xlabel='WEALTH Quartile Number for < median age',  yticks = np.arange(0, 5 , 1))
ax1.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

fig, ax2 = plt.subplots(figsize=(6, 6))

sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==1)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 1')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==2)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 2')
sns.regplot (scatter = False, data = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)].reset_index(drop = True), x = 'Quartile Number', y = df4.loc[(df4['age'] >=age_m) & (df4['isced']==3)]['Bayesianreasoning_1'], fit_reg=True, ci=None, ax=ax2, label='ISCED = 3')

ax2.errorbar(x = df4.loc[ (df4['age'] >= age_m)  ]["Quartile Number"].unique(), y = df4.loc[ (df4['age'] >=age_m)  ].groupby(['Quartile Number'])['Bayesianreasoning_1'].mean(), yerr = df4.loc[ (df4['age'] >=age_m) ].groupby(['Quartile Number'])['Bayesianreasoning_1'].sem(), fmt='o', color = lighten_color('gray',0.5))
ax2.set(ylabel='Bayesianreasoning_1', xlabel='WEALTH Quartile Number for >=median age',  yticks = np.arange(0, 5 , 1))
ax2.legend(loc='center left',bbox_to_anchor=(1.0, 0.5))
plt.show()

	ResponseId	Please indicate your Gender.	Please mark your age (in years)	What is currently your highest Education?	Please state your current occupation.	How do you describe your willingness to take financial risk in general?	Given the number of years that you have held various investments and the amount of investing you might have done, what degree of investment experience in the stock market do you have?	If an expert tries to worry or scare me, i.e. a financial advisor about my financial situation, I choose another expert.	I only buy a financial product I understand.	I trust doctors.	When I want to buy a bigger item like a refrigerator or an expensive item of clothing, I wait a month to see whether I still want it and only buy it then.	I always keep in mind that everything I do on the web could be used to my disadvantage.	In my household, we/I spend:	In my household, we/I	_Distribution 1_\n\nHow risky do you perceive the investment to be?	_Distribution 2_\n\nHow risky do you perceive the investment to be?	_Distribution 3_\n\nHow risky do you perceive the investment to be?	_Distribution 4_\n\nHow risky do you perceive the investment to be?	_Distribution 5_\n\nHow risky do you perceive the investment to be?	Distribution 6\n\nHow risky do you perceive the investment to be?	Distribution 7\n\nHow risky do you perceive the investment to be?	Distribution 8\n\nHow risky do you perceive the investment to be?	Mumbai A = 9 out of 10000 OR B = 1 out of 1000	Bengaluru A = 0.7% OR B = 0.099%	Kolkata A = 0.61% OR B = 6 out of 10000	HIV test	Fingerprint	DNA test	Cancer screening test	Professional horoscope	A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%. Which of the following statements is true?	There is an official prediction that the national stock market will grow 2% annually over the next 5 years. This means that…	Imagine you are told that the price of the stock Soya Ruchi increases from INR 60 to INR 120 after the company merger. What does this mean?	It is predicted that Indigo Bank has 30% chance of default next year. Which of the following alternatives is the most appropriate interpretation of the statement?	The probability that the economy will go into a recession this year is 30%. If the economy goes into recession, the probability that the stock market will decrease is 80%. If the economy does not go into a recession, the probability that the stock market will decrease is 23%. What is the probability that the economy goes into recession given that the stock market decreased?	A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:	Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000. This implies:	Imagine that we flip a fair coin 1,000 times. What is your best guess about how many times the coin will come up heads in 1,000 flips? \n\n\_\_\_\_\_\_ times out of 1,000.	In the Bingo Lottery, the chance of winning a $10 prize is 1%. What is your best guess about how many people will win a $10 prize if 1,000 people each buy a single ticket for Bingo Lottery?\n\n\_\_\_\_\_\_ person(s) out of 1,000.	In a sweepstakes, the chance of winning a car is 1 in 1,000. What percentage of tickets for the sweepstakes wins a car?\n\n\_\_\_\_\_ % of tickets	About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]	Approximately what percentage (%) of people who die from cancer die from colon cancer, breast cancer, and prostate cancer taken together?	The following figure shows the number of men and women among a group of smartphone users. The total number of circles is 100. \n\nHow many more men than women are there among the 100 people using a smartphone?	In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?	Please indicate your approximate annual personal income from all sources for last year	Please provide a rough guess (in Indian Rupees) of the worth of your household's assets. Please do not forget to correct it for your debts, such as a mortgage or any loans you might have.	uid	Response Type	Start Date (UTC)	Stage Date (UTC)	Submit Date (UTC)	Network ID	LTA
2	6458w9qj8sqw7ldtt56457hzeduqz2qb	Female	36 - 45	Post Graduate	Employee/Consultant in Other than the Finance ...	6	10	Completely	Just about	Completely	Just about	Completely	all or more than the household income, even th...	have an emergency fund, and spend as it feels ...	7	3	6	7	6	7	6	7	B	A	B	Yes	Yes	Yes	Yes	Yes	The lower the quality of the study, the more l...	the growth rate over five years will be betwee...	the stock price increased by 100%	30% of central bankers think that Indigo Bank ...	80%	100 in 10,000 people prior to the intervention...	The medication increases recovery by 100%	10	100	10	59 out of 100	20	10	Crosicol	35,00,001 - 45,00,000	10	MXzTR1725956643Vmr	completed	2024-09-10 08:25:58	NaN	2024-09-10 08:33:57	942e517f31	5.543981
3	9woo3kquplxhob9whhbfvhleezgy3m8l	Male	25 - 35	Post Graduate	Employee/Consultant in Other than the Finance ...	5	3	Somewhat	Somewhat	Somewhat	Not at all	Somewhat	all or more than our household income, because...	write down budget and spending, and have an em...	4	2	4	1	1	7	6	3	A	B	A	Yes	Yes	Yes	No	No	Irrespective of the quality of the study, futu...	the growth rate over five years will be betwee...	the stock price increased by 50%	Banks similar to Indigo will default 30% of th...	30%	5 in 100 people prior to the intervention to 6...	The medication increases recovery by 2%	500	10	10	59 out of 1000	50	10	Hertinol	5,00,000 - 15,00,000	50000	MXhzL1725956594UqF	completed	2024-09-10 08:25:48	NaN	2024-09-10 08:35:16	dcc2ff7969	6.574074
4	bago2g5wtk2g5u2h0xbago2g00a6b3jx	Female	18 - 25	Post Graduate	Student	3	2	Somewhat	Just about	Completely	Completely	Just about	less than half of the household income and sav...	spend as it feel right, and do not have an eme...	3	4	6	3	1	7	7	5	B	A	B	Yes	Yes	Yes	No	No	The higher the quality of the study, the more ...	the growth rate will be 0.4% on average each year	the stock price increased by 50%	The bank will default on 30% of repayments in ...	80%	it is not possible to determine which of the a...	The medication increases recovery by 100%	500	10	10	9 out of 59	25	60	Hertinol	5,00,000 - 15,00,000	4000000	MXE5b1725956481kmw	completed	2024-09-10 08:25:12	NaN	2024-09-10 08:36:22	ed3799240f	7.754630
5	huy0uxnebqrbqdegrz2sxhofkhuy0ux0	Female	25 - 35	Under Graduate	Home Maker or not employed	5	0	Somewhat	Moderately	Moderately	Somewhat	Not at all	all or more than the household income, even th...	have an emergency fund, and spend as it feels ...	5	4	3	6	1	2	7	3	A	A	B	Yes	Yes	Yes	Yes	Yes	The higher the quality of the study, the more ...	the growth rate over five years will be exactl...	the stock price increased by 100%	30% of the banks customers will default next year	30%	5 in 100 people prior to the intervention to 6...	None of the above is implied.	20	1	1	59 out of 100	50	40	Can’t say	< 5,00,000	500000	MXWUf1725956389QtP	completed	2024-09-10 08:25:04	NaN	2024-09-10 08:34:00	b9b7c6efb9	6.203704
7	bm66xvxw1reh3fbm668aomqq4gckt8ao	Male	25 - 35	Post Graduate	Employee/Consultant in Other than the Finance ...	7	10	Completely	Completely	Completely	Completely	Completely	all or more than the household income, even th...	have an emergency fund, and spend as it feels ...	7	7	7	7	7	7	7	7	A	A	A	No	Yes	Yes	No	Yes	The higher the quality of the study, the more ...	the growth rate over five years will be exactl...	the stock price increased by 100%	Banks similar to Indigo will default 30% of th...	50%	100 in 10,000 people prior to the intervention...	The medication increases recovery by 50%	800	500	60	59 out of 1000	30	50	Crosicol	> 45,00,000	400000	MXnNH1725956452gcb	completed	2024-09-10 08:24:01	NaN	2024-09-10 08:37:41	097676ebd8	9.490741
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
635	7my3wgtsbfwf8wr4f90f7my3wy5i8juf	Male	56 - 65	Under Graduate	Employee/Consultant in Other than the Finance ...	6	7	Just about	Just about	Just about	Just about	Just about	all or more than our household income, because...	write down budget and spending, but have no em...	5	5	5	5	5	5	6	6	B	B	A	No	Yes	No	No	Yes	Irrespective of the quality of the study, futu...	the growth rate over five years will be betwee...	the stock price increased by 50%	30% of the banks customers will default next year	50%	5 in 100 people prior to the intervention to 6...	The medication increases recovery by 100%	1000	1000	100	9 out of 10	50	50	Crosicol	5,00,000 - 15,00,000	999900	MXX1J1724688179GOm	completed	2024-08-26 16:04:51	NaN	2024-08-26 16:14:59	4ae93bad32	7.037037
637	ukj87cqk7sxt1a977ukj7dlqa19b34pg	Female	25 - 35	Post Graduate	Employee/Consultant in Other than the Finance ...	4	5	Somewhat	Moderately	Moderately	Not at all	Somewhat	more than half of the household income and sav...	write down budget and spending, and have an em...	6	6	6	2	3	1	3	5	A	A	B	Yes	Yes	Yes	Yes	Yes	Irrespective of the quality of the study, futu...	the growth rate over five years will be betwee...	the stock price increased by 50%	30% of the banks customers will default next year	30%	it is not possible to determine which of the a...	The medication increases recovery by 2%	500	250	50	9 out of 10	90	60	Can’t say	< 5,00,000	100000	MXxJi1724688182ARM	completed	2024-08-26 16:04:22	NaN	2024-08-26 16:13:02	ff51233502	6.018519
638	tod55nr27ozkyci5tod5rup87g8lo9ss	Male	18 - 25	Diploma or vocation training	Entrepreneur or Own Business	4	2	Somewhat	Moderately	Moderately	Somewhat	Moderately	all or more than our household income, because...	write down budget and spending, and have an em...	1	4	4	2	5	5	0	4	B	B	A	Yes	Yes	Yes	No	No	Irrespective of the quality of the study, futu...	the growth rate will be 0.4% on average each year	an answer is not possible based on the informa...	The bank will default on 30% of repayments in ...	50%	70 in 100 people prior to the intervention to ...	None of the above is implied.	500	10	1	59 out of 100	50	10	Hertinol	25,00,001 - 35,00,000	5000000	MXLz61724688143Vic	completed	2024-08-26 16:04:07	NaN	2024-08-26 16:12:43	e7e6df96c7	5.972222
641	13nxdmew77mtq97scc6r13nxduhpev0l	Male	25 - 35	Post Graduate	Entrepreneur or Own Business	5	9	Moderately	Moderately	Moderately	Moderately	Somewhat	less than half of the household income and sav...	have an emergency fund, and spend as it feels ...	3	3	5	1	0	3	7	2	A	B	A	Yes	Yes	Yes	Yes	No	The higher the quality of the study, the more ...	the growth rate will be 0.4% on average each year	the stock price increased by 100%	30% of the banks customers will default next year	30%	100 in 10,000 people prior to the intervention...	The medication increases recovery by 50%	500	100	1	59 out of 1000	25	60	Crosicol	15,00,001 - 25,00,000	13000000	MXcTq1724679467kFu	completed	2024-08-26 13:39:40	NaN	2024-08-26 13:49:13	5a7cce9582	6.631944
642	s5m2omt6lyzdyeb1k9cs5m2omtmv05c3	Female	18 - 25	Post Graduate	Employee/Consultant in Other than the Finance ...	6	1	Somewhat	Moderately	Moderately	Moderately	Moderately	more than half of the household income and sav...	write down budget and spending, and have an em...	5	0	6	1	6	6	0	6	A	A	A	Yes	Yes	Yes	Yes	No	The lower the quality of the study, the more l...	it is not possible to predict the growth rate ...	the stock price increased by 50%	The bank will default on 30% of repayments in ...	30%	70 in 100 people prior to the intervention to ...	The medication increases recovery by 50%	500	10	1	59 out of 1000	25	60	Crosicol	15,00,001 - 25,00,000	100000	MX4Er1724679488ZwW	completed	2024-08-26 13:39:25	NaN	2024-08-26 13:54:38	3e73e75dfb	10.567130

		ResponseId	Please indicate your Gender.	Please mark your age (in years)	What is currently your highest Education?	Please state your current occupation.	How do you describe your willingness to take financial risk in general?	Given the number of years that you have held various investments and the amount of investing you might have done, what degree of investment experience in the stock market do you have?	If an expert tries to worry or scare me, i.e. a financial advisor about my financial situation, I choose another expert.	I only buy a financial product I understand.	I trust doctors.	When I want to buy a bigger item like a refrigerator or an expensive item of clothing, I wait a month to see whether I still want it and only buy it then.	I always keep in mind that everything I do on the web could be used to my disadvantage.	In my household, we/I spend:	In my household, we/I	_Distribution 1_\n\nHow risky do you perceive the investment to be?	_Distribution 2_\n\nHow risky do you perceive the investment to be?	_Distribution 3_\n\nHow risky do you perceive the investment to be?	_Distribution 4_\n\nHow risky do you perceive the investment to be?	_Distribution 5_\n\nHow risky do you perceive the investment to be?	Distribution 6\n\nHow risky do you perceive the investment to be?	Distribution 7\n\nHow risky do you perceive the investment to be?	Distribution 8\n\nHow risky do you perceive the investment to be?	Mumbai A = 9 out of 10000 OR B = 1 out of 1000	Bengaluru A = 0.7% OR B = 0.099%	Kolkata A = 0.61% OR B = 6 out of 10000	HIV test	Fingerprint	DNA test	Cancer screening test	Professional horoscope	A study estimates that eating 100g chocolate everyday increases the risk of obesity by 20%. Which of the following statements is true?	There is an official prediction that the national stock market will grow 2% annually over the next 5 years. This means that…	Imagine you are told that the price of the stock Soya Ruchi increases from INR 60 to INR 120 after the company merger. What does this mean?	It is predicted that Indigo Bank has 30% chance of default next year. Which of the following alternatives is the most appropriate interpretation of the statement?	The probability that the economy will go into a recession this year is 30%. If the economy goes into recession, the probability that the stock market will decrease is 80%. If the economy does not go into a recession, the probability that the stock market will decrease is 23%. What is the probability that the economy goes into recession given that the stock market decreased?	A new policy intervention increases the number of people who are employed by 20%. This statistic implies that the intervention increases the number of people who are employed from:	Imagine you are told that a new medication increases the number of people who recover from a disease from 2 out of 1,000 to 4 out of 1,000. This implies:	Imagine that we flip a fair coin 1,000 times. What is your best guess about how many times the coin will come up heads in 1,000 flips? \n\n\_\_\_\_\_\_ times out of 1,000.	In the Bingo Lottery, the chance of winning a $10 prize is 1%. What is your best guess about how many people will win a $10 prize if 1,000 people each buy a single ticket for Bingo Lottery?\n\n\_\_\_\_\_\_ person(s) out of 1,000.	In a sweepstakes, the chance of winning a car is 1 in 1,000. What percentage of tickets for the sweepstakes wins a car?\n\n\_\_\_\_\_ % of tickets	About 10 out of 1,000 children develop Down syndrome. In a Down syndrome test, 9 out of these 10 children with Down syndrome tested positive. Out of the 990 children without Down syndrome 50 nevertheless tested positive. Among those women with a positive test result concerning their child how many actually have a child with Down syndrome? [Select one response only]	Approximately what percentage (%) of people who die from cancer die from colon cancer, breast cancer, and prostate cancer taken together?	The following figure shows the number of men and women among a group of smartphone users. The total number of circles is 100. \n\nHow many more men than women are there among the 100 people using a smartphone?	In a magazine you see two advertisements, one on page 5 and another on page 12. Each is for a different drug for treating heart disease, and each includes a graph showing the effectiveness of the drug compared to a placebo (sugar pill).\n\nCompared to the placebo, which treatment leads to a larger decrease in the percentage of patients who die?	Please indicate your approximate annual personal income from all sources for last year	Please provide a rough guess (in Indian Rupees) of the worth of your household's assets. Please do not forget to correct it for your debts, such as a mortgage or any loans you might have.	uid	Response Type	Start Date (UTC)	Stage Date (UTC)	Submit Date (UTC)	Network ID	LTA	age_rec	age	isced	income	wealth	q8_2_1	q8_2_2	q8_2_3	q8_2_4	q8_2_5	q8_3	q8_4	q8_5berlin_1	q8_5london_1	q8_5paris_1	q8_6	q8_7	q9_1_1	q9_2_1	q9_3	q10_1_1	q10_2_1	q10_3_1	q10_4	certainty1	certainty2	certainty3	certainty4	certainty5	uncertainty1	uncertainty2	numeracy1	numeracy2	numeracy3	numeracy4	numeracy5	graph1	graph2	graph3	riskcomprehension1	riskcomprehension2	riskcomprehension3	riskcomprehension4	riskcomprehension5	bayesian1
ResponseId
00ujdxbfoya0donu8r00ujcjdkojc99x	428	00ujdxbfoya0donu8r00ujcjdkojc99x	Female	25 - 35	Under Graduate	Employee/Consultant in Other than the Finance ...	7	8	Completely	Completely	Completely	Moderately	Moderately	less than half of the household income and sav...	write down budget and spending, and have an em...	2	1	1	1	0	2	0	0	B	B	A	Yes	Yes	Yes	Yes	No	The higher the quality of the study, the more ...	the growth rate over five years will be betwee...	the stock price increased by 50%	The bank will default on 30% of repayments in ...	50%	100 in 10,000 people prior to the intervention...	The medication increases recovery by 50%	500	100	1	59 out of 1000	50	10	They are equal	< 5,00,000	1500000	MXic21725178744djB	completed	2024-09-01 08:23:04	NaN	2024-09-01 09:24:44	1c1d51349c	42.824074	2	30.5	2	1	1500000	1	1	1	1	2	2	3	2	2	1	2	2	50	10	3	500	100	1	3	0	0	0	0	1	0	0	1	0	0	0	0	0	0	1	1	0	1	0	0	0
01ro2iftnum5v4cn01r03c75aa24emhq	528	01ro2iftnum5v4cn01r03c75aa24emhq	Female	18 - 25	Under Graduate	Student	5	8	Just about	Completely	Completely	Completely	Just about	more than half of the household income and sav...	write down budget and spending, and have an em...	6	6	2	6	6	6	2	4	A	B	B	Yes	Yes	No	Yes	Yes	The lower the quality of the study, the more l...	the growth rate over five years will be exactl...	an answer is not possible based on the informa...	Banks similar to Indigo will default 30% of th...	50%	70 in 100 people prior to the intervention to ...	The medication increases recovery by 100%	46	100	99	59 out of 1000	79	40	Hertinol	5,00,000 - 15,00,000	500000	MX7wI1724844822zNm	completed	2024-08-28 11:36:50	NaN	2024-08-28 11:51:07	81a8a6836a	9.918981	2	21.5	2	2	500000	1	1	2	1	1	1	2	1	2	2	1	3	79	40	2	46	100	99	3	0	0	1	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0
022xoawhrghfhv4a1g022xmz7hb0m41f	442	022xoawhrghfhv4a1g022xmz7hb0m41f	Female	46 - 55	Under Graduate	Entrepreneur or Own Business	5	4	Somewhat	Somewhat	Moderately	Just about	Somewhat	all or more than the household income, even th...	spend as it feel right, and do not have an eme...	4	4	4	4	4	4	4	4	B	A	B	Yes	Yes	Yes	No	No	The lower the quality of the study, the more l...	the growth rate will be 0.4% on average each year	the stock price increased by 60%	30% of the banks customers will default next year	80%	5 in 100 people prior to the intervention to 6...	The medication increases recovery by 2%	500	1	1	59 out of 1000	17	65	They are equal	< 5,00,000	15000000	MXFDH1725178659xRn	completed	2024-09-01 08:21:14	NaN	2024-09-01 08:33:49	fbd8962401	8.738426	3	50.5	2	1	15000000	1	1	1	2	2	1	1	2	1	2	3	1	17	65	3	500	1	1	3	0	0	0	1	1	1	0	1	0	0	0	0	0	0	1	1	1	0	0	0	0
02pxtdbyibecqqfvwlw02pxwfbane9zd	558	02pxtdbyibecqqfvwlw02pxwfbane9zd	Female	46 - 55	Post Graduate	Employee/Consultant in Other than the Finance ...	6	10	Not at all	Completely	Completely	Completely	Moderately	all or more than the household income, even th...	have an emergency fund, and spend as it feels ...	5	5	6	5	7	5	0	5	B	A	A	Yes	Yes	Yes	No	Yes	The higher the quality of the study, the more ...	it is not possible to predict the growth rate ...	an answer is not possible based on the informa...	30% of the banks customers will default next year	30%	5 in 100 people prior to the intervention to 6...	None of the above is implied.	1000	1	100	9 out of 10	20	20	Can’t say	5,00,000 - 15,00,000	3000000	MXIpM1724844742xet	completed	2024-08-28 11:35:05	NaN	2024-08-28 11:48:47	e8b668d11b	9.513889	3	50.5	1	2	3000000	1	1	1	2	1	2	4	2	1	1	4	1	20	20	4	1000	1	100	2	0	0	0	1	0	0	1	0	0	0	0	0	0	1	0	1	1	1	0	0	0
037aefjdt26mnjd102nz0pk037aedfhc	544	037aefjdt26mnjd102nz0pk037aedfhc	Male	46 - 55	Post Graduate	Employee/Consultant in Other than the Finance ...	3	6	Somewhat	Somewhat	Somewhat	Somewhat	Somewhat	more than half of the household income and sav...	write down budget and spending, but have no em...	3	1	6	1	2	6	7	5	A	B	A	No	No	Yes	No	No	Irrespective of the quality of the study, futu...	it is not possible to predict the growth rate ...	an answer is not possible based on the informa...	Banks similar to Indigo will default 30% of th...	60%	70 in 100 people prior to the intervention to ...	None of the above is implied.	1000	1	10	9 out of 59	50	20	Hertinol	5,00,000 - 15,00,000	7000000	MXpv61724844846UVs	completed	2024-08-28 11:35:50	NaN	2024-08-28 11:45:54	f51b723011	6.990741	3	50.5	1	2	7000000	2	2	1	2	2	3	4	1	2	1	4	3	50	20	2	1000	1	10	1	1	1	0	1	1	0	1	0	0	0	0	0	0	1	0	0	0	1	0	0	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
zcj7ldokhyo6217f9sriwizcj7ldodwt	268	zcj7ldokhyo6217f9sriwizcj7ldodwt	Female	36 - 45	Under Graduate	Entrepreneur or Own Business	5	8	Completely	Completely	Completely	Not at all	Moderately	less than half of the household income and sav...	write down budget and spending, and have an em...	2	0	7	5	5	7	0	4	A	B	A	No	Yes	Yes	No	No	The higher the quality of the study, the more ...	it is not possible to predict the growth rate ...	the stock price increased by 100%	Banks similar to Indigo will default 30% of th...	80%	it is not possible to determine which of the a...	The medication increases recovery by 2%	500	10	1	9 out of 59	25	60	Hertinol	15,00,001 - 25,00,000	600000000	MX6PC1725259690mDV	completed	2024-09-02 06:53:29	NaN	2024-09-02 07:28:48	237c2e5291	24.525463	3	40.5	2	3	600000000	2	1	1	2	2	2	4	1	2	1	3	4	25	60	2	500	10	1	1	1	0	0	1	1	0	1	1	1	0	0	0	1	0	0	0	0	1	0	1	1
zfemo30rg0ekva18x1kjwzfemo2wpg2p	185	zfemo30rg0ekva18x1kjwzfemo2wpg2p	Male	36 - 45	Ph.D. or higher	Employee/Consultant in Other than the Finance ...	6	8	Not at all	Moderately	Just about	Not at all	Not at all	less than half of the household income and sav...	have an emergency fund, and spend as it feels ...	3	1	4	2	0	5	0	2	B	A	A	No	Yes	Yes	No	No	Irrespective of the quality of the study, futu...	the growth rate over five years will be betwee...	the stock price increased by 100%	30% of central bankers think that Indigo Bank ...	30%	100 in 10,000 people prior to the intervention...	The medication increases recovery by 100%	500	10	1	59 out of 1000	25	20	Can’t say	15,00,001 - 25,00,000	8500000	MXUxt1725260651heG	completed	2024-09-02 07:08:18	NaN	2024-09-02 07:23:09	fbb956cffc	10.312500	3	40.5	1	3	8500000	2	1	1	2	2	4	3	2	1	1	1	2	25	20	4	500	10	1	3	1	0	0	1	1	0	0	1	1	0	0	0	1	1	0	1	1	1	1	0	0
znqvw4t38br3072znqvwvjsgb6wvj7nt	305	znqvw4t38br3072znqvwvjsgb6wvj7nt	Female	36 - 45	Post Graduate	Entrepreneur or Own Business	5	5	Somewhat	Completely	Just about	Moderately	Somewhat	less than half of the household income and sav...	spend as it feel right, and do not have an eme...	0	0	0	1	0	0	0	0	B	B	B	Yes	Yes	Yes	No	No	Irrespective of the quality of the study, futu...	it is not possible to predict the growth rate ...	the stock price increased by 50%	30% of the banks customers will default next year	80%	it is not possible to determine which of the a...	The medication increases recovery by 2%	500	10	1	9 out of 59	25	20	Hertinol	5,00,000 - 15,00,000	2000000	MXd921725259652XDJ	completed	2024-09-02 06:49:56	NaN	2024-09-02 07:17:20	1964f49e1c	19.027778	3	40.5	1	2	2000000	1	1	1	2	2	3	4	2	2	2	3	4	25	20	2	500	10	1	1	0	0	0	1	1	0	1	1	1	0	0	0	1	1	0	1	0	0	0	1	1
zstpq23h1x3ilab8s8vsrbkw8t4zstpq	306	zstpq23h1x3ilab8s8vsrbkw8t4zstpq	Male	18 - 25	Under Graduate	Entrepreneur or Own Business	6	9	Completely	Completely	Moderately	Completely	Moderately	all or more than our household income, because...	have an emergency fund, and spend as it feels ...	6	3	7	1	2	7	7	3	A	A	B	No	Yes	Yes	No	No	The higher the quality of the study, the more ...	the growth rate over five years will be betwee...	the stock price increased by 60%	Banks similar to Indigo will default 30% of th...	80%	5 in 100 people prior to the intervention to 6...	None of the above is implied.	750	10	1	59 out of 1000	50	10	They are equal	5,00,000 - 15,00,000	60000000	MXfmg1725259705Fkj	completed	2024-09-02 06:49:52	NaN	2024-09-02 07:01:04	03aa77ec48	7.777778	2	21.5	2	2	60000000	2	1	1	2	2	2	3	1	1	2	4	1	50	10	3	750	10	1	3	1	0	0	1	1	0	0	0	1	0	0	0	0	0	1	0	1	0	0	0	0
ztfdlcbiyo1xco3ztfdl4ozqpte2q3m9	95	ztfdlcbiyo1xco3ztfdl4ozqpte2q3m9	Male	25 - 35	Post Graduate	Entrepreneur or Own Business	6	7	Somewhat	Just about	Completely	Moderately	Somewhat	more than half of the household income and sav...	spend as it feel right, and do not have an eme...	6	0	2	5	3	7	0	4	B	A	A	No	Yes	Yes	Yes	Yes	The higher the quality of the study, the more ...	it is not possible to predict the growth rate ...	the stock price increased by 100%	30% of central bankers think that Indigo Bank ...	80%	5 in 100 people prior to the intervention to 6...	The medication increases recovery by 2%	500	10	1	59 out of 1000	25	60	Can’t say	15,00,001 - 25,00,000	5000000	MXjp51725955476hMt	completed	2024-09-10 08:06:34	NaN	2024-09-10 08:16:35	f68294192e	6.956019	2	30.5	1	3	5000000	2	1	1	1	1	2	4	2	1	1	3	1	25	60	4	500	10	1	3	1	0	0	0	0	0	1	1	1	0	0	0	1	0	0	1	1	1	0	0	0

		ResponseId	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
ResponseId
00ujdxbfoya0donu8r00ujcjdkojc99x	428	00ujdxbfoya0donu8r00ujcjdkojc99x	30.5	2	2	1	1500000	1	0	2	1	1	0	20.0	0.0	40.0	33.333333	50.0	0.0	5	27.777778
01ro2iftnum5v4cn01r03c75aa24emhq	528	01ro2iftnum5v4cn01r03c75aa24emhq	21.5	2	2	2	500000	1	1	1	0	0	0	20.0	50.0	20.0	0.000000	0.0	0.0	3	16.666667
022xoawhrghfhv4a1g022xmz7hb0m41f	442	022xoawhrghfhv4a1g022xmz7hb0m41f	50.5	3	2	1	15000000	2	1	2	1	1	0	40.0	50.0	40.0	33.333333	50.0	0.0	7	38.888889
02pxtdbyibecqqfvwlw02pxwfbane9zd	558	02pxtdbyibecqqfvwlw02pxwfbane9zd	50.5	3	1	2	3000000	1	1	3	1	0	0	20.0	50.0	60.0	33.333333	0.0	0.0	6	33.333333
037aefjdt26mnjd102nz0pk037aedfhc	544	037aefjdt26mnjd102nz0pk037aedfhc	50.5	3	1	2	7000000	4	1	1	1	0	1	80.0	50.0	20.0	33.333333	0.0	100.0	8	44.444444
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
zcj7ldokhyo6217f9sriwizcj7ldodwt	268	zcj7ldokhyo6217f9sriwizcj7ldodwt	40.5	3	2	3	600000000	3	1	2	1	2	1	60.0	50.0	40.0	33.333333	100.0	100.0	10	55.555556
zfemo30rg0ekva18x1kjwzfemo2wpg2p	185	zfemo30rg0ekva18x1kjwzfemo2wpg2p	40.5	3	1	3	8500000	3	0	4	2	2	0	60.0	0.0	80.0	66.666667	100.0	0.0	11	61.111111
znqvw4t38br3072znqvwvjsgb6wvj7nt	305	znqvw4t38br3072znqvwvjsgb6wvj7nt	40.5	3	1	2	2000000	2	1	2	2	2	1	40.0	50.0	40.0	66.666667	100.0	100.0	10	55.555556
zstpq23h1x3ilab8s8vsrbkw8t4zstpq	306	zstpq23h1x3ilab8s8vsrbkw8t4zstpq	21.5	2	2	2	60000000	3	0	1	1	1	0	60.0	0.0	20.0	33.333333	50.0	0.0	6	33.333333
ztfdlcbiyo1xco3ztfdl4ozqpte2q3m9	95	ztfdlcbiyo1xco3ztfdl4ozqpte2q3m9	30.5	2	1	3	5000000	1	1	3	1	2	0	20.0	50.0	60.0	33.333333	100.0	0.0	8	44.444444

	q8_2_1	q8_2_2	q8_2_3	q8_2_4	q8_2_5	q8_3	q8_4	q8_5berlin_1	q8_5london_1	q8_5paris_1	q8_6	q8_7	q9_1_1	q9_2_1	q9_3	q10_1_1	q10_2_1	q10_3_1	q10_4
0	1	2	2	2	1	4	3	2	2	2	1	1	25.0	20	3	500	10	0.10	2
1	1	1	1	2	2	4	4	2	2	2	1	2	25.0	20	4	500	10	0.10	3
2	2	1	2	2	2	2	4	1	1	1	2	3	50.0	20	3	1000	10	50.00	1
3	2	1	1	2	2	2	2	1	1	1	2	2	25.0	20	3	500	10	0.10	3
4	1	1	1	1	2	2	2	2	2	2	1	2	25.0	20	3	500	10	0.10	3
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
365	2	2	2	2	1	3	2	1	1	1	3	3	50.0	40	3	500	68	0.01	4
366	2	2	2	1	2	2	2	2	2	2	3	4	25.0	20	3	500	50	0.10	1
367	2	2	1	2	2	3	4	1	1	1	3	4	90.0	20	3	345	1	0.10	2
368	1	1	1	1	2	2	3	2	2	2	1	3	25.0	5	2	600	10	0.10	2
369	2	2	1	2	2	2	4	2	2	2	1	1	25.0	10	2	500	500	50.00	3

		ResponseId	Q0	Q1	Q2	Q3	Q4	Q5_1	Q6_1	Q8_1	Q8_2	Q8_3	Q8_4	Q8_5	Q8_6	Q8_7	Q8_8	Q9b_1	Q9b_4	Q9b_5	Q9b_6	Q9b_7	Q9b_8	Q9b_9	Q11a_1	Q11a_2	Q11a_3	Q11b	Q11c	Q11d	Q11h	Attention Check	Q11i	Q12a	Q12b	Q12c	Q13a	Q13b	Q13c	Q13d	Q14a	Q14b	Q14c	Q15a	Q14b.1	age	age_rec	isced	income	wealth	q8_2_1	q8_2_2	q8_2_3	q8_2_4	q8_2_5	q8_3	q8_4	q8_5berlin_1	q8_5london_1	q8_5paris_1	q8_6	q8_7	q9_1_1	q9_2_1	q9_3	q10_1_1	q10_2_1	q10_3_1	q10_4	certainty1	certainty2	certainty3	certainty4	certainty5	uncertainty1	uncertainty2	numeracy1	numeracy2	numeracy3	numeracy4	numeracy5	graph1	graph2	graph3	riskcomprehension1	riskcomprehension2	riskcomprehension3	riskcomprehension4	riskcomprehension5	bayesian1
ResponseId
R_1YkMM2lMB9aEuVL	150	R_1YkMM2lMB9aEuVL	Yes, I would like to participate in the study ...	Female	41	Undergraduate Program	Salaried/Employee/Consultant in a sector other...	3	3	3	4	3	3	3	3	3	4	85.0	0.0	0	0	5.0	10	0.0	1	1	1	DNA test	The higher the quality of the study, the more ...	The growth rate over five years will be betwee...	5 in 100 people prior to the intervention to 6...	Vase	The medication increases recovery by 50%	Less than $102	Exactly the same as today with the money in th...	True	1000	10	1.0	59 out of 1000	9.0	20	Crosicol	< INR 500,000	25000.0	41	3	2	1	25000.0	2	2	1	2	2	2	3	1	1	1	2	1	9.0	20	1	1000	10	1.0	3	1	1	0	1	1	0	0	0	1	0	0	0	0	1	0	0	1	1	0	0	0
R_40N8bUn3C0jwGzf	200	R_40N8bUn3C0jwGzf	Yes, I would like to participate in the study ...	Female	18	Undergraduate Program	Student	0 (unwilling to take risk)	0 (no investment experience)	3	3	4	3	3	3	2	3	40.0	20.0	10	10	10.0	5	5.0	2	1	1	HIV test,Fingerprint,DNA test,Cancer screening...	The higher the quality of the study, the more ...	The growth rate over five years will be betwee...	It is not possible to determine which of the a...	Vase	The medication increases recovery by 2%	More than $102	More than today with the money in this account	True	1000	500	100.0	9 out of 59	11.0	60	They are equal	< INR 500,000	80000.0	18	2	2	1	80000.0	1	1	1	1	2	2	3	2	2	2	3	4	11.0	60	3	1000	500	100.0	1	0	0	0	0	1	0	0	0	0	0	0	0	0	0	1	1	0	0	0	1	1
R_40VyTmJ6i96wUP4	235	R_40VyTmJ6i96wUP4	Yes, I would like to participate in the study ...	Female	34	Post-Graduate Program	Not employed	2	0 (no investment experience)	3	2	2	3	3	3	2	3	40.0	25.0	10	10	15.0	0	0.0	1	2	1	DNA test	The higher the quality of the study, the more ...	It is not possible to predict the growth rate ...	It is not possible to determine which of the a...	Vase	The medication increases recovery by 2%	Refuse to answer	Less than today with the money in this account	Do not know	698	352	1.0	59 out of 1000	30.0	20	Crosicol	< INR 500,000	500000.0	34	2	1	1	500000.0	2	2	1	2	2	2	4	1	1	1	3	4	30.0	20	1	698	352	1.0	3	1	1	0	1	1	0	1	0	0	0	0	0	0	1	0	0	1	1	0	1	0
R_40ZEg7vX3Y7mMQl	184	R_40ZEg7vX3Y7mMQl	Yes, I would like to participate in the study ...	Female	31	Undergraduate Program	Not employed	4	6	2	2	2	1 (strongly disagree)	2	3	4	1 (strongly disagree)	30.0	30.0	10	0	20.0	10	0.0	1	1	1	HIV test,Fingerprint,DNA test,Cancer screening...	The higher the quality of the study, the more ...	The growth rate will be 0.4% on average each year	70 in 100 people prior to the intervention to ...	Vase	The medication increases recovery by 2%	More than $102	Less than today with the money in this account	True	500	10	10.0	59 out of 1000	50.0	20	They are equal	< INR 500,000	500000.0	31	2	2	1	500000.0	1	1	1	1	2	2	1	1	1	1	3	3	50.0	20	3	500	10	10.0	3	0	0	0	0	1	0	0	1	1	0	0	0	0	1	1	0	1	1	0	0	0
R_40cbsHWzTyKyFxv	53	R_40cbsHWzTyKyFxv	Yes, I would like to participate in the study ...	Male	38	Undergraduate Program	Salaried/Employee/Consultant in a sector other...	5	5	4	6 (strongly agree)	5	4	6 (strongly agree)	6 (strongly agree)	6 (strongly agree)	6 (strongly agree)	45.0	35.0	0	5	15.0	0	0.0	2	1	2	Fingerprint	The higher the quality of the study, the more ...	The growth rate over five years will be exactl...	It is not possible to determine which of the a...	Vase	The medication increases recovery by 2%	More than $102	Exactly the same as today with the money in th...	False	1000	100	25.0	9 out of 10	30.0	20	They are equal	< INR 500,000	300000.0	38	3	2	1	300000.0	2	1	2	2	2	2	2	2	2	2	3	4	30.0	20	3	1000	100	25.0	2	1	0	1	1	1	0	0	0	0	0	0	0	0	1	1	1	0	0	0	1	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
R_4yzFrFYMa7w8nQZ	238	R_4yzFrFYMa7w8nQZ	Yes, I would like to participate in the study ...	Female	18	Secondary School (11th to 12th Std.)	Student	0 (unwilling to take risk)	0 (no investment experience)	6 (strongly agree)	5	5	5	4	4	5	6 (strongly agree)	20.0	10.0	15	25	5.0	9	16.0	1	2	1	DNA test	The higher the quality of the study, the more ...	The growth rate over five years will be exactl...	70 in 100 people prior to the intervention to ...	Vase	The medication increases recovery by 100%	More than $102	Less than today with the money in this account	True	1000	100	100.0	59 out of 1000	25.0	20	Hertinol	< INR 500,000	90000.0	18	2	3	1	90000.0	2	2	1	2	2	2	2	1	1	1	1	3	25.0	20	2	1000	100	100.0	3	1	1	0	1	1	0	0	0	0	0	0	0	1	1	0	0	1	1	1	0	0
R_4zPiSk23ayGvFD4	364	R_4zPiSk23ayGvFD4	Yes, I would like to participate in the study ...	Male	70	Undergraduate Program	Salaried/Employee/Consultant in a sector other...	4	6	4	4	4	4	5	5	5	4	15.0	20.0	10	0	25.0	30	0.0	2	1	1	DNA test	The higher the quality of the study, the more ...	It is not possible to predict the growth rate ...	It is not possible to determine which of the a...	Vase	The medication increases recovery by 100%	More than $102	Less than today with the money in this account	False	500	10	1.0	59 out of 1000	25.0	20	Crosicol	INR 1500,001 – INR 30,00,000	7000000.0	70	4	2	3	7000000.0	2	2	1	2	2	2	4	2	2	2	1	4	25.0	20	1	500	10	1.0	3	1	1	0	1	1	0	1	1	1	0	0	0	1	1	0	1	0	0	1	1	0
R_4zdzjECYeO2A5Ox	255	R_4zdzjECYeO2A5Ox	Yes, I would like to participate in the study ...	Female	19	Secondary School (11th to 12th Std.)	Student	4	2	6 (strongly agree)	4	5	5	5	6 (strongly agree)	3	4	40.0	50.0	0	10	0.0	0	0.0	2	1	1	Fingerprint,DNA test	The lower the quality of the study, the more l...	The growth rate over five years will be exactl...	70 in 100 people prior to the intervention to ...	Vase	None of the above is implied	More than $102	Less than today with the money in this account	Do not know	500	10	0.1	9 out of 59	25.0	20	They are equal	< INR 500,000	800000.0	19	2	3	1	800000.0	2	1	1	2	2	1	2	2	2	2	4	3	25.0	20	3	500	10	0.1	1	1	0	0	1	1	1	0	1	1	1	0	0	1	1	1	1	0	0	0	0	1
R_8k6D0jzzHCC5X3Z	369	R_8k6D0jzzHCC5X3Z	Yes, I would like to participate in the study ...	Male	25	Undergraduate Program	Salaried/Employee/Consultant in a sector other...	7 (willing to take risk)	8	6 (strongly agree)	5	5	4	6 (strongly agree)	6 (strongly agree)	6 (strongly agree)	5	5.0	5.0	5	5	30.0	20	30.0	2	1	1	DNA test	The higher the quality of the study, the more ...	It is not possible to predict the growth rate ...	5 in 100 people prior to the intervention to 6...	Vase	The medication increases recovery by 100%	More than $102	Less than today with the money in this account	False	500	500	50.0	59 out of 1000	25.0	10	Hertinol	> INR 75,00,000	8000000.0	25	2	2	5	8000000.0	2	2	1	2	2	2	4	2	2	2	1	1	25.0	10	2	500	500	50.0	3	1	1	0	1	1	0	1	1	0	0	0	0	1	0	0	1	0	0	1	0	0
R_8taN3wUPbeZMxzM	2	R_8taN3wUPbeZMxzM	Yes, I would like to participate in the study ...	Male	20	Undergraduate Program	Student	3	0 (no investment experience)	6 (strongly agree)	6 (strongly agree)	4	1 (strongly disagree)	1 (strongly disagree)	2	1 (strongly disagree)	6 (strongly agree)	30.0	40.0	10	10	0.0	5	5.0	1	1	1	Fingerprint	The higher the quality of the study, the more ...	It is not possible to predict the growth rate ...	70 in 100 people prior to the intervention to ...	Vase	The medication increases recovery by 50%	More than $102	More than today with the money in this account	Do not know	1000	10	50.0	9 out of 59	50.0	20	They are equal	< INR 500,000	10000.0	20	2	2	1	10000.0	2	1	2	2	2	2	4	1	1	1	2	3	50.0	20	3	1000	10	50.0	1	1	0	1	1	1	0	1	0	1	0	0	0	0	1	1	0	1	1	0	0	1

DATA CLEANING AND WRANGLING¶

Typeform Data¶

Demographics¶

Questions¶

Scoring¶

BeSample Data¶

Scoring¶

GRAPHS¶

ISCED¶

INCOME¶

INCOME VS SCORES¶

INCOME vs SCORES w/ ISCED classification¶

INCOME vs SCORES w/ ISCED and AGE based classification¶

WEALTH¶

WEALTH vs SCORES w/ ISCED classification¶

WEALTH Quartile Number vs SCORES w/ ISCED and AGE based classification¶

	ResponseId	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
0	00ujdxbfoya0donu8r00ujcjdkojc99x	30.5	2	2	1	1500000.0	1	0	2	1	1	0	20.0	0.0	40.0	33.333333	50.0	0.0	5	27.777778
1	01ro2iftnum5v4cn01r03c75aa24emhq	21.5	2	2	2	500000.0	1	1	1	0	0	0	20.0	50.0	20.0	0.000000	0.0	0.0	3	16.666667
2	022xoawhrghfhv4a1g022xmz7hb0m41f	50.5	3	2	1	15000000.0	2	1	2	1	1	0	40.0	50.0	40.0	33.333333	50.0	0.0	7	38.888889
3	02pxtdbyibecqqfvwlw02pxwfbane9zd	50.5	3	1	2	3000000.0	1	1	3	1	0	0	20.0	50.0	60.0	33.333333	0.0	0.0	6	33.333333
4	037aefjdt26mnjd102nz0pk037aedfhc	50.5	3	1	2	7000000.0	4	1	1	1	0	1	80.0	50.0	20.0	33.333333	0.0	100.0	8	44.444444
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
834	R_4yzFrFYMa7w8nQZ	18.0	2	3	1	90000.0	4	0	3	2	0	0	80.0	0.0	60.0	66.666667	0.0	0.0	9	50.000000
835	R_4zPiSk23ayGvFD4	70.0	4	2	3	7000000.0	4	1	3	2	2	0	80.0	50.0	60.0	66.666667	100.0	0.0	12	66.666667
836	R_4zdzjECYeO2A5Ox	19.0	2	3	1	800000.0	3	1	1	3	2	1	60.0	50.0	20.0	100.000000	100.0	100.0	11	61.111111
837	R_8k6D0jzzHCC5X3Z	25.0	2	2	5	8000000.0	4	1	2	1	1	0	80.0	50.0	40.0	33.333333	50.0	0.0	9	50.000000
838	R_8taN3wUPbeZMxzM	20.0	2	2	1	10000.0	4	1	2	2	1	1	80.0	50.0	40.0	66.666667	50.0	100.0	11	61.111111

	ResponseId	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
0	R_4dyDxqwOf6c5bUt	46.0	3	1	2	4000000.0	4	1	1	2	2	1	80.0	50.0	20.0	66.666667	100.0	100.0	11	61.111111
1	R_4VgFPHXva4gJjhf	21.0	2	1	2	900000.0	4	0	1	2	0	0	80.0	0.0	20.0	66.666667	0.0	0.0	7	38.888889
2	R_4Vwgb0WSlLsIYlL	44.0	3	1	1	10000000.0	4	1	3	1	1	0	80.0	50.0	60.0	33.333333	50.0	0.0	10	55.555556
3	p15xolgvzls8wctop15q6ithwa3t8msq	40.5	3	1	3	50000000.0	0	0	2	3	1	0	0.0	0.0	40.0	100.000000	50.0	0.0	6	33.333333
4	oza83gum5po6u3jk4oza83ad7gtuqvzd	30.5	2	1	4	150000.0	1	0	1	0	1	1	20.0	0.0	20.0	0.000000	50.0	100.0	4	22.222222
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
834	R_42AoOsYAIUhnViQ	35.0	2	3	3	5000.0	4	1	2	2	0	0	80.0	50.0	40.0	66.666667	0.0	0.0	9	50.000000
835	870h2qs77chp1cd5q870n4fnkjbrxezk	21.5	2	3	2	150000.0	2	0	0	0	0	1	40.0	0.0	0.0	0.000000	0.0	100.0	3	16.666667
836	8eynniv5jpifaxbqslk8eynnw7he4d4q	40.5	3	3	1	200000.0	1	2	1	0	0	0	20.0	100.0	20.0	0.000000	0.0	0.0	4	22.222222
837	sw8rv2orxir5asm6jn8sw8rv24olibmz	21.5	2	3	1	30000.0	1	0	3	0	1	0	20.0	0.0	60.0	0.000000	50.0	0.0	5	27.777778
838	R_4dE2pi4mUJFadlk	18.0	2	3	1	50000.0	4	2	3	3	0	0	80.0	100.0	60.0	100.000000	0.0	0.0	12	66.666667

	Certainty_%	Uncertainty_%	RiskComprehension_%	Numeracy_%	GraphLiteracy_%	Bayesianreasoning_%
isced
1	44.987893	22.760291	44.116223	53.874092	44.552058	19.612591
2	52.747253	22.390110	44.285714	54.532967	45.879121	25.549451
3	62.903226	33.870968	41.612903	48.387097	44.086022	22.580645

	Certainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_%
income
1	58.053691	42.885906	48.210291	53.020134	25.167785	46.290082
2	44.802432	44.316109	44.984802	55.927052	21.276596	42.046606
3	45.818182	47.818182	44.545455	53.636364	20.000000	42.979798
4	43.076923	42.769231	37.948718	50.769231	21.538462	39.572650
5	48.648649	41.081081	35.135135	45.945946	18.918919	39.189189

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	413.000000	413.000000	413.0	413.000000	4.130000e+02	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000	413.000000
mean	35.285714	2.443099	1.0	2.314770	1.020335e+07	2.249395	0.455206	2.205811	1.336562	1.077482	0.196126	44.987893	22.760291	44.116223	44.552058	53.874092	19.612591	7.520581	41.781006
std	9.828602	0.582747	0.0	1.091761	3.794182e+07	1.293479	0.616168	1.069781	0.963179	0.771961	0.397546	25.869582	30.808415	21.395627	32.105960	38.598049	39.754648	2.523229	14.017942
min	18.000000	2.000000	1.0	1.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	2.000000	11.111111
25%	30.500000	2.000000	1.0	2.000000	3.000000e+05	1.000000	0.000000	1.000000	1.000000	0.000000	0.000000	20.000000	0.000000	20.000000	33.333333	0.000000	0.000000	6.000000	33.333333
50%	30.500000	2.000000	1.0	2.000000	1.200000e+06	2.000000	0.000000	2.000000	1.000000	1.000000	0.000000	40.000000	0.000000	40.000000	33.333333	50.000000	0.000000	7.000000	38.888889
75%	40.500000	3.000000	1.0	3.000000	5.000000e+06	3.000000	1.000000	3.000000	2.000000	2.000000	0.000000	60.000000	50.000000	60.000000	66.666667	100.000000	0.000000	9.000000	50.000000
max	70.500000	4.000000	1.0	5.000000	5.000000e+08	5.000000	2.000000	5.000000	3.000000	2.000000	1.000000	100.000000	100.000000	100.000000	100.000000	100.000000	100.000000	15.000000	83.333333

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	364.000000	364.000000	364.0	364.000000	3.640000e+02	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000	364.000000
mean	30.877747	2.271978	2.0	1.851648	9.861171e+06	2.637363	0.447802	2.214286	1.376374	1.090659	0.255495	52.747253	22.390110	44.285714	45.879121	54.532967	25.549451	8.021978	44.566545
std	10.056769	0.514458	0.0	1.017734	5.225548e+07	1.323794	0.612110	1.019487	0.995080	0.823231	0.436739	26.475871	30.605488	20.389748	33.169331	41.161531	43.673914	2.646180	14.701002
min	18.000000	2.000000	2.0	1.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	1.000000	5.555556
25%	22.000000	2.000000	2.0	1.000000	2.000000e+05	1.000000	0.000000	2.000000	1.000000	0.000000	0.000000	20.000000	0.000000	40.000000	33.333333	0.000000	0.000000	6.000000	33.333333
50%	30.500000	2.000000	2.0	2.000000	6.000000e+05	3.000000	0.000000	2.000000	1.000000	1.000000	0.000000	60.000000	0.000000	40.000000	33.333333	50.000000	0.000000	8.000000	44.444444
75%	34.000000	2.000000	2.0	2.000000	3.000000e+06	4.000000	1.000000	3.000000	2.000000	2.000000	1.000000	80.000000	50.000000	60.000000	66.666667	100.000000	100.000000	10.000000	55.555556
max	70.000000	4.000000	2.0	5.000000	6.000000e+08	5.000000	2.000000	5.000000	3.000000	2.000000	1.000000	100.000000	100.000000	100.000000	100.000000	100.000000	100.000000	15.000000	83.333333

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	62.000000	62.000000	62.0	62.000000	6.200000e+01	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000	62.000000
mean	26.951613	2.225806	3.0	1.629032	3.719399e+06	3.145161	0.677419	2.080645	1.322581	0.967742	0.225806	62.903226	33.870968	41.612903	44.086022	48.387097	22.580645	8.419355	46.774194
std	12.274093	0.584481	0.0	1.119629	1.336024e+07	1.157255	0.672022	0.892563	0.971296	0.829136	0.421526	23.145094	33.601075	17.851257	32.376538	41.456813	42.152552	2.583354	14.351966
min	16.000000	0.000000	3.0	1.000000	0.000000e+00	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000	20.000000	0.000000	0.000000	0.000000	0.000000	0.000000	3.000000	16.666667
25%	19.000000	2.000000	3.0	1.000000	5.000000e+04	2.000000	0.000000	2.000000	1.000000	0.000000	0.000000	40.000000	0.000000	40.000000	33.333333	0.000000	0.000000	7.000000	38.888889
50%	21.250000	2.000000	3.0	1.000000	2.000000e+05	4.000000	1.000000	2.000000	1.000000	1.000000	0.000000	80.000000	50.000000	40.000000	33.333333	50.000000	0.000000	9.000000	50.000000
75%	30.500000	2.000000	3.0	2.000000	1.500000e+06	4.000000	1.000000	3.000000	2.000000	2.000000	0.000000	80.000000	50.000000	60.000000	66.666667	100.000000	0.000000	10.000000	55.555556
max	70.500000	4.000000	3.0	5.000000	1.000000e+08	5.000000	2.000000	4.000000	3.000000	2.000000	1.000000	100.000000	100.000000	80.000000	100.000000	100.000000	100.000000	12.000000	66.666667

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	573.000000	573.0	573.000000	573.000000	5.730000e+02	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000	573.000000
mean	26.943281	2.0	1.647469	1.959860	7.616613e+06	2.554974	0.476440	2.158813	1.371728	1.085515	0.235602	51.099476	23.821990	43.176265	45.724258	54.275742	23.560209	7.883072	43.794842
std	4.633303	0.0	0.626887	1.047872	3.540355e+07	1.350679	0.638141	1.016084	0.973263	0.805151	0.424745	27.013572	31.907039	20.321679	32.442106	40.257572	42.474533	2.596786	14.426588
min	18.000000	2.0	1.000000	1.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	1.000000	5.555556
25%	22.000000	2.0	1.000000	1.000000	2.000000e+05	1.000000	0.000000	1.000000	1.000000	0.000000	0.000000	20.000000	0.000000	20.000000	33.333333	0.000000	0.000000	6.000000	33.333333
50%	30.000000	2.0	2.000000	2.000000	6.500000e+05	3.000000	0.000000	2.000000	1.000000	1.000000	0.000000	60.000000	0.000000	40.000000	33.333333	50.000000	0.000000	8.000000	44.444444
75%	30.500000	2.0	2.000000	2.000000	3.000000e+06	4.000000	1.000000	3.000000	2.000000	2.000000	0.000000	80.000000	50.000000	60.000000	66.666667	100.000000	0.000000	10.000000	55.555556
max	35.000000	2.0	3.000000	5.000000	5.000000e+08	5.000000	2.000000	5.000000	3.000000	2.000000	1.000000	100.000000	100.000000	100.000000	100.000000	100.000000	100.000000	15.000000	83.333333

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	232.000000	232.0	232.000000	232.000000	2.320000e+02	232.000000	232.000000	232.000000	232.000000	232.000000	232.000000	232.000000	232.000000	232.000000	232.000000	232.00000	232.000000	232.000000	232.000000
mean	42.892241	3.0	1.426724	2.306034	1.445481e+07	2.314655	0.448276	2.271552	1.284483	1.047414	0.198276	46.293103	22.413793	45.431034	42.816092	52.37069	19.827586	7.564655	42.025862
std	4.753812	0.0	0.591258	1.138221	6.124404e+07	1.282635	0.578771	1.048380	0.987419	0.790853	0.399563	25.652698	28.938528	20.967600	32.913966	39.54263	39.956313	2.591219	14.395662
min	36.000000	3.0	1.000000	1.000000	3.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.00000	0.000000	2.000000	11.111111
25%	40.500000	3.0	1.000000	1.750000	3.000000e+05	1.000000	0.000000	2.000000	0.000000	0.000000	0.000000	20.000000	0.000000	40.000000	0.000000	0.00000	0.000000	6.000000	33.333333
50%	40.500000	3.0	1.000000	2.000000	1.750000e+06	2.000000	0.000000	2.000000	1.000000	1.000000	0.000000	40.000000	0.000000	40.000000	33.333333	50.00000	0.000000	8.000000	44.444444
75%	46.500000	3.0	2.000000	3.000000	7.500000e+06	4.000000	1.000000	3.000000	2.000000	2.000000	0.000000	80.000000	50.000000	60.000000	66.666667	100.00000	0.000000	9.000000	50.000000
max	54.000000	3.0	3.000000	5.000000	6.000000e+08	5.000000	2.000000	5.000000	3.000000	2.000000	1.000000	100.000000	100.000000	100.000000	100.000000	100.00000	100.000000	14.000000	77.777778

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	33.000000	33.0	33.000000	33.000000	3.300000e+01	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000	33.000000
mean	62.969697	4.0	1.484848	2.181818	9.582270e+06	2.393939	0.454545	2.393939	1.545455	1.090909	0.212121	47.878788	22.727273	47.878788	51.515152	54.545455	21.212121	8.090909	44.949495
std	4.390934	0.0	0.618527	1.236288	1.406214e+07	0.966288	0.616994	1.248484	0.938446	0.765001	0.415149	19.325756	30.849709	24.969679	31.281550	38.250074	41.514875	2.602446	14.458036
min	57.000000	4.0	1.000000	1.000000	1.000000e+05	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000	20.000000	0.000000	0.000000	0.000000	0.000000	0.000000	3.000000	16.666667
25%	60.500000	4.0	1.000000	1.000000	1.250000e+06	2.000000	0.000000	1.000000	1.000000	1.000000	0.000000	40.000000	0.000000	20.000000	33.333333	50.000000	0.000000	6.000000	33.333333
50%	60.500000	4.0	1.000000	2.000000	5.000000e+06	2.000000	0.000000	2.000000	2.000000	1.000000	0.000000	40.000000	0.000000	40.000000	66.666667	50.000000	0.000000	8.000000	44.444444
75%	65.000000	4.0	2.000000	2.000000	1.000000e+07	3.000000	1.000000	3.000000	2.000000	2.000000	0.000000	60.000000	50.000000	60.000000	66.666667	100.000000	0.000000	10.000000	55.555556
max	70.500000	4.0	3.000000	5.000000	5.250000e+07	4.000000	2.000000	5.000000	3.000000	2.000000	1.000000	80.000000	100.000000	100.000000	100.000000	100.000000	100.000000	13.000000	72.222222

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	298.000000	298.000000	298.000000	298.000000	2.980000e+02	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000	298.000000
mean	23.206376	1.993289	1.869128	1.788591	7.382464e+06	3.036913	0.553691	2.171141	1.459732	1.151007	0.271812	60.738255	27.684564	43.422819	48.657718	57.550336	27.181208	8.644295	48.023863
std	3.435092	0.115857	0.635366	1.050383	3.704148e+07	1.248021	0.665759	0.939718	0.984761	0.804459	0.445642	24.960412	33.287967	18.794362	32.825354	40.222937	44.564190	2.504068	13.911487
min	16.000000	0.000000	1.000000	1.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	2.000000	11.111111
25%	21.000000	2.000000	1.000000	1.000000	1.000000e+05	2.000000	0.000000	2.000000	1.000000	0.000000	0.000000	40.000000	0.000000	40.000000	33.333333	0.000000	0.000000	7.000000	38.888889
50%	22.000000	2.000000	2.000000	1.000000	5.000000e+05	4.000000	0.000000	2.000000	1.000000	1.000000	0.000000	80.000000	0.000000	40.000000	33.333333	50.000000	0.000000	9.000000	50.000000
75%	26.000000	2.000000	2.000000	2.000000	2.000000e+06	4.000000	1.000000	3.000000	2.000000	2.000000	1.000000	80.000000	50.000000	60.000000	66.666667	100.000000	100.000000	10.000000	55.555556
max	30.000000	2.000000	3.000000	5.000000	5.000000e+08	4.000000	2.000000	5.000000	3.000000	2.000000	1.000000	80.000000	100.000000	100.000000	100.000000	100.000000	100.000000	15.000000	83.333333

	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%
count	541.000000	541.000000	541.000000	541.000000	5.410000e+02	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000	541.000000
mean	38.018484	2.550832	1.423290	2.214418	1.078387e+07	2.179298	0.421442	2.216266	1.293900	1.033272	0.197782	43.585952	21.072089	44.325323	43.130006	51.663586	19.778189	7.341959	40.788663
std	9.235180	0.608356	0.561193	1.084510	4.690035e+07	1.263159	0.589902	1.084996	0.968265	0.792793	0.398695	25.263180	29.495078	21.699921	32.275495	39.639645	39.869541	2.530458	14.058100
min	30.500000	2.000000	1.000000	1.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	1.000000	5.555556
25%	30.500000	2.000000	1.000000	1.000000	3.000000e+05	1.000000	0.000000	1.000000	1.000000	0.000000	0.000000	20.000000	0.000000	20.000000	33.333333	0.000000	0.000000	6.000000	33.333333
50%	35.000000	2.000000	1.000000	2.000000	1.200000e+06	2.000000	0.000000	2.000000	1.000000	1.000000	0.000000	40.000000	0.000000	40.000000	33.333333	50.000000	0.000000	7.000000	38.888889
75%	40.500000	3.000000	2.000000	3.000000	5.000000e+06	3.000000	1.000000	3.000000	2.000000	2.000000	0.000000	60.000000	50.000000	60.000000	66.666667	100.000000	0.000000	9.000000	50.000000
max	70.500000	4.000000	3.000000	5.000000	6.000000e+08	5.000000	2.000000	5.000000	3.000000	2.000000	1.000000	100.000000	100.000000	100.000000	100.000000	100.000000	100.000000	14.000000	77.777778

	ResponseId	age	age_rec	isced	income	wealth	Certainty_5	Uncertainty_2	RiskComprehension_5	GraphLiteracy_3	Numeracy_2	Bayesianreasoning_1	Certainty_%	Uncertainty_%	RiskComprehension_%	GraphLiteracy_%	Numeracy_%	Bayesianreasoning_%	TotalScore_18	TotalScore_%	Quartile Number
0	R_4MJAzsNYYA69Y8p	27.0	2	3	1	0.0	4	1	3	1	0	1	80.0	50.0	60.0	33.333333	0.0	100.0	10	55.555556	1
1	R_41bfnamM0zpH94i	30.0	2	1	2	0.0	4	0	2	1	1	1	80.0	0.0	40.0	33.333333	50.0	100.0	9	50.000000	1
2	e7ty2tbwa1hcmte7ty683aw64pla7689	30.5	2	2	1	0.0	1	0	2	1	1	0	20.0	0.0	40.0	33.333333	50.0	0.0	5	27.777778	1
3	R_4lnmGavSf6rw1eE	31.0	2	1	1	0.0	1	0	2	1	2	0	20.0	0.0	40.0	33.333333	100.0	0.0	6	33.333333	1
4	R_4DPGUOAP86yO30R	19.0	2	3	1	0.0	4	2	1	2	1	1	80.0	100.0	20.0	66.666667	50.0	100.0	11	61.111111	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
834	lmoqk1qcf44jqqudwkplmoqk18cao0fs	40.5	3	1	2	400000000.0	2	0	1	1	1	0	40.0	0.0	20.0	33.333333	50.0	0.0	5	27.777778	4
835	wn9mk6m1k8o10l1twn9mk6b4z7l7yvhc	50.5	3	1	3	500000000.0	2	0	3	3	2	0	40.0	0.0	60.0	100.000000	100.0	0.0	10	55.555556	4
836	oczyh23wpxpk4o0teoczy5zzzng3i6kk	30.5	2	2	5	500000000.0	1	0	3	3	2	0	20.0	0.0	60.0	100.000000	100.0	0.0	9	50.000000	4
837	35zux4sc18rplyz3dc85z35zux4sdfm4	21.5	2	2	4	500000000.0	2	0	1	0	1	0	40.0	0.0	20.0	0.000000	50.0	0.0	4	22.222222	4
838	zcj7ldokhyo6217f9sriwizcj7ldodwt	40.5	3	2	3	600000000.0	3	1	2	1	2	1	60.0	50.0	40.0	33.333333	100.0	100.0	10	55.555556	4